Collective alternative dispute resolution
(ADR) for the private enforcement of EU
competition law
Egelyn Braun
Thesis submitted for assessment with a view to obtaining the degree of
Master in Comparative, European and International Laws (LL.M.)
of the European University Institute
Florence, 2016
European University Institute
Department of Law
Collective alternative dispute resolution (ADR) for the private
enforcement of EU competition law
Egelyn Braun
Thesis submitted for assessment with a view to obtaining the degree of
Master in Comparative, European and International Laws (LL.M.)
of the European University Institute
Supervisor
Giorgio Monti, European University Institute
© Braun, 2016
No part of this thesis may be copied, reproduced or transmitted without prior
permission of the author
Researcher declaration to accompany the submission of written work
Department of Law – LL.M. Programme
I, Egelyn Braun, certify that I am the author of the work ‘Collective alternative dispute
resolution (ADR) for the private enforcement of EU competition law’ I have presented for
examination for the LL.M. at the European University Institute. I also certify that this is
solely my own original work, other than where I have clearly indicated, in this declaration and
in the thesis, that it is the work of others.
I warrant that I have obtained all the permissions required for using any material from other
copyrighted publications.
I certify that this work complies with the Code of Ethics in Academic Research issued by the
European University Institute (IUE 332/2/10 (CA 297).
The copyright of this work rests with its author. Quotation from this thesis is permitted,
provided that full acknowledgement is made. This work may not be reproduced without my
prior written consent. This authorisation does not, to the best of my knowledge, infringe the
rights of any third party.
I declare that this work, with footnotes, consists of 37 754 words.
01.09.2016
Summary
The European enforcement landscape is undergoing significant changes that are leading to a
departure from the actors, tools and processes traditionally associated with delivering justice.
This thesis examines these themes while developing a solution to the private enforcement gap
that continues to leave a large number of victims without a remedy, particularly if they have
suffered low-value individual harm as a result of competition infringements. In order to
ensure that the private enforcement of EU competition law leads to the effective enforcement
of EU rights and to the full compensation of all victims, a collective redress device must be
developed. In particular, this thesis will explore whether optimal private enforcement
outcomes could be achieved through the integration of collective alternative dispute
resolution (‘collective ADR’) into a regulatory enforcement architecture as a first choice
redress avenue. To date, the use of collective ADR as a private enforcement mechanism has
not been considered as a serious policy option on the European level. While this thesis
focuses on the use of collective ADR in the context of competition enforcement, it also
confronts issues that could be expanded to private enforcement in other fields. Ultimately, the
enforcement toolbox should be diversified not only to ensure the successful fulfilment of the
regulatory goals, but also to facilitate the transformations that are occurring in the
enforcement landscape more broadly.
Table of Contents
Introduction .............................................................................................................................1
Chapter 1: Setting the scene...................................................................................................3
1.1 What is collective ADR? .................................................................................................3
1.2 Why competition law? .....................................................................................................5
1.3 A toolbox of collective redress ........................................................................................6
1.4 EU policy developments ..................................................................................................9
Legislative saga ..............................................................................................................9
Prevailing discourse .....................................................................................................10
1.5 Collective ADR in EU law ............................................................................................12
Damages Directive ........................................................................................................13
Recommendation on Collective Redress .......................................................................15
Conclusion ....................................................................................................................16
Chapter 2: Incentives ............................................................................................................19
2.1 Infringing undertakings .................................................................................................19
Undertakings with economic, social and normative motivations .................................20
Case study: Lufthansa and the air cargo cartel ............................................................23
2.2 Victims and their representatives ...................................................................................25
2.3 Designing a regulatory enforcement architecture ..........................................................28
Responsive private enforcement: carrots and sticks .....................................................30
Chapter 3: Safeguards ..........................................................................................................35
3.1 When, how and under what circumstances....................................................................35
Consenting to collective ADR .......................................................................................38
Case study: class arbitration ........................................................................................41
3.2 Balancing incentives and safeguards .............................................................................43
Once-and-for-all: recognition and enforcement ...........................................................44
Fairness review .............................................................................................................46
Chapter 4: A comparative analysis of collective ADR in the Netherlands and the UK ..51
4.1 The Netherlands .............................................................................................................51
4.1.1 Toolbox of collective proceedings ..........................................................................51
Collective litigation.......................................................................................................52
Collective settlements....................................................................................................52
4.1.2 Public enforcer capacities .......................................................................................56
4.2 The United Kingdom .....................................................................................................58
4.2.1 Toolbox of collective proceedings ..........................................................................58
Collective litigation.......................................................................................................59
Collective settlements....................................................................................................61
Voluntary redress schemes ............................................................................................62
4.2.2 Public enforcer capacities .......................................................................................65
4.3 Comparative analysis .....................................................................................................67
Overview .......................................................................................................................69
(1) Structural incentives for collective ADR .................................................................71
(2) Funding of representative entities ...........................................................................72
(3) Structural differences ..............................................................................................73
(4) Fairness review .......................................................................................................75
(5) Role of the public enforcer ......................................................................................77
Conclusion ....................................................................................................................78
Chapter 5: The implications of integrating collective ADR into EU competition
enforcement ...........................................................................................................................81
5.1 Impact on the public-private interplay and the goals of enforcement ...........................81
Public-private transformations .....................................................................................81
The impact on deterrence ..............................................................................................83
The impact on compensation ........................................................................................85
Deterrence, compensation and ‘resolution’ ..................................................................87
5.2 Collective ADR in the Commission’s public enforcement policy .................................89
Treatment of voluntary compensation...........................................................................90
Availability of negotiation procedures ..........................................................................92
Treatment of self-regulatory compliance ......................................................................94
Conclusion .............................................................................................................................97
Bibliography ........................................................................................................................101
Introduction
The European enforcement landscape is undergoing significant changes that are leading to a
departure from the actors, tools and processes traditionally associated with delivering justice.
These changes include the increasing collectivisation and privatisation of enforcement, as
well as the abandonment of the binary distinction between public and private enforcement.1
Against this background, it is unfortunate that the development of a European collective
redress device has not yet culminated with a legislative instrument, leaving many victims
without an adequate remedy.2 In particular, there is an identifiable enforcement gap for largescale low-value damage resulting from competition infringements, which is predominantly
suffered by consumers. In order to ensure that the private enforcement of EU competition law
leads to the effective enforcement of EU rights and to the full compensation of all victims, a
collective redress device must be provided. Ultimately, the enforcement toolbox should be
diversified not only to ensure the successful fulfilment of the regulatory goals, but also to
facilitate the transformations that are occurring in the enforcement landscape more broadly.
This thesis examines whether it would be possible to achieve private enforcement outcomes
for large groups of victims that are currently without a remedy in the most optimal manner
with the consent of all parties involved. It is submitted that such outcomes can be achieved
through the integration of collective alternative dispute resolution (‘collective ADR’) into a
regulatory enforcement architecture as a first choice redress avenue. To date, the use of
collective ADR as a private enforcement mechanism has not been considered as a serious
policy option on the European level. The thesis explores the collective ADR proposal in the
context of competition enforcement, however, it also confronts issues that could be expanded
to private enforcement in other fields.
This thesis is structured as follows. The first chapter sets the scene by defining collective
ADR, its relationship with collective litigation, the policy developments on the EU level
1
H.-W. Micklitz & A. Wechsler (eds), The Transformation of Enforcement: European Economic Law in Global
Perspective (Hart Publishing 2016).
2
A decade of policy developments culminated with a non-binding instrument; Commission Recommendation of
11 June 2013 on common principles for injunctive and compensatory collective redress mechanisms in the
Member States concerning violations of rights granted under Union Law, OJ L 201. A collective redress device
was also notably absent from the competition damages directive; Directive 2014/104/EU of the European
Parliament and of the Council of 26 November 2014 on certain rules governing actions for damages under
national law for infringements of the competition law provisions of the Member States and of the European
Union, OJ L 349.
1
regarding collective redress and private enforcement, and the presence of collective ADR in
existing legislation, particularly in the Damages Directive. The second and third chapter will
start to construct the regulatory enforcement architecture that would be necessary for the
successful integration of collective ADR. In particular, the second chapter inspects the
incentives that would have to be created for the infringing undertakings, victims and their
representatives in order to consensually initiate collective ADR, after which the third chapter
considers the safeguards that would have to be created in order to ensure fairness guarantees.
After outlining the theoretical framework, the fourth chapter presents a comparative analysis
of two collective ADR models in the national enforcement architectures of the UK and the
Netherlands in order to illustrate how collective ADR works in practice and to draw lessons
for the development of a European model. The fifth chapter reflects on the proposal more
broadly, particularly on its ability to meet the goals of private enforcement and its
transformative implications for the European Commission’s current practice. The thesis
concludes with preliminary policy recommendations for the development of a collective ADR
model.
2
Chapter 1: Setting the scene
1.1 What is collective ADR?
Collective ADR is an umbrella term that will be used to describe voluntary alternative-tocourt dispute resolution processes that can be used in collective harm cases in order to reach a
consensual outcome. Foremost, collective ADR is a voluntary and consensual process. The
expression of consent can occur at different stages – for instance, the initiation of ADR is
generally voluntary for all parties involved, however, if national civil justice systems made
ADR a mandatory step in civil proceedings, then compliance with the ADR outcome would
nevertheless be voluntary. Furthermore, ADR processes are generally confidential, subject to
some legal and practical limitations, particularly in case a large number of parties are
involved. The collective dimension would start from two or more victims and include both
legal and natural persons. Moreover, ADR processes could function with an intermediary
public or private entity, or without the involvement of any third parties. Third party
involvement could take several forms, ranging from simply facilitating negotiations to issuing
a binding decision. If the ADR process is successful, then it could lead to a binding or nonbinding outcome. A binding agreement could take the form of a settlement, whereas a binding
decision could be any arrangement that was accepted by the parties prior to the initiation of a
third party procedure (e.g. arbitration, which leads to an arbitral award). Finally, in contrast
with court judgments, there is generally more flexibility as to the substantive content of the
ADR outcomes.
The current landscape of ADR mechanisms and entities in the Member States is highly
multifaceted and new forms of offline and online dispute resolution are in continuous
development. It is therefore important not to exclude any innovative forms of ADR through
the use of rigid definitions. In order to give a systematic account of the different forms of
collective ADR that exist, the following categorization could be developed. The categories are
presented in a hierarchical order according to the level of formalism and bindingness that they
feature:
Collective negotiation without an ADR entity - The most informal type of collective ADR is
direct negotiation aimed at reaching a consensual agreement in the form of a settlement.
Direct negotiations are generally conducted without the use of ADR entities and rely on the
3
legal representatives of the parties. The direct negotiation category could also include various
internal complaint mechanisms, which the undertaking has developed.
Collective negotiation through an ADR entity - The key examples of negotiation-based forms
of ADR are mediation and conciliation, in which a third party actively assists the parties in
coming to a consensual agreement, but does not issue a binding decision. With conciliation,
the third party also makes a suggestion for terms of the settlement or issues a non-binding
opinion.
Collective adjudication through an ADR entity - ADR entities other than arbitral tribunals can
offer adjudicative procedures, in which a third party issues a decision, which is binding
through contracts. However, such decisions will generally provide less certainty in terms of
recognition and enforcement. Moreover, ADR entities across the EU have limited experience
with mass claims.3
Collective Ombudsman procedures through an ADR entity - A public or private Ombudsman
is an independent third party, who can employ various ADR mechanisms, including
adjudicative procedures or mediation, which result in a binding or non-binding decision.
Ombudsmen are often created as sector-specific ADR entities. As there are many procedural
variations in existence, the definition of Ombudsman is far from harmonized.
Collective arbitration through an arbitration tribunal - The most court-like type of collective
ADR is arbitration, in which an adjudicating third party issues a decision - the arbitral award that is binding on the basis of contract law and international agreements. Due to its formal
adjudicative nature, combined with strong recognition and enforcement effects, arbitration is
sometimes distinguished from other forms of ADR. Moreover, unlike other ADR entities,
arbitration tribunals do have considerable experience with managing mass claims, which are
often referred to as multi-party or class arbitration, depending on their opt-in or opt-out
nature.
3
For a comparative study on consumer ADR entities and mass torts, see C. Hodges & A. Stadler (eds),
Resolving Mass Disputes: ADR and Settlement of Mass Claims (Edward Elgar Publishing 2013).
4
Having obtained a basic overview of ADR mechanisms and entities, the next preliminary
point regards the context in which collective ADR will be placed in this thesis - the private
enforcement of EU competition law.
1.2 Why competition law?
To begin with, the EU enjoys a high level of competence in competition matters, which has
allowed for the harmonization of substantive competition law. As demonstrated by the rapid
development of private enforcement on the EU level, substantive harmonization has made the
harmonization of procedural rules significantly easier to justify.4 A significant step towards
harmonizing competition-specific tort rules has already been taken with the adoption of the
Damages Directive. 5 However, despite these developments, the current framework is
inadequate for the weakest victims of competition infringements.
In particular, collective proceedings are still missing from the private enforcement toolbox
and additional measures must be envisioned in order to ensure the ‘effective enforcement of
EU rights’ and ‘full compensation’ for all victims. The presence of an identifiable
enforcement gap for large-scale low-value damage sets EU competition law apart from many
other fields. Currently, private enforcement claims are only being brought by a specific group
of undertakings: competitors and direct purchasers. According to research conducted in the
UK, Germany, France, Italy and Spain, barely any cases have been brought by indirect
purchasers and none by umbrella purchasers.6 One of the primary reasons for this dynamic is
the fact that competitors and direct purchasers generally have the sufficient resources to bring
actions and due to their close connection to the infringement, their claims are substantively
and procedurally easier to establish. It is submitted that this enforcement gap could be closed
by introducing a collective redress device.
Moreover, both anecdotal and empirical evidence would suggest that a large share of EU
competition law disputes never reach courts because they are settled. 7 For instance,
M. Ioannidou, Consumer Involvement in Private EU Competition Law Enforcement (2nd edn, Oxford
University Press 2015), p. 189.
5
Damages Directive (n 2).
6
I. Lianos, P. Davis & P. Nebbia, Damages Claims for the Infringement of EU Competition Law (Oxford
University Press 2015), p. 73.
7
D. Waelbroeck, D. Slater & G. Even-Shoshan, ‘Ashurst Study on the Conditions of Claims for Damages in
Case of Infringement of EC Competition Rules’, Brussels, 31 August, 2004, p. 134; for a UK study, see B.
4
5
settlements could be preferred in order to reduce costs and inconvenience, as well as to
maintain existing relationships and safeguard one’s reputation. If the out-of-court settlement
of claims is already the hidden reality of private enforcement, then targeted regulation could
be envisioned, for instance in order to ensure fairness guarantees for the weaker parties.
Finally, the existence of powerful public enforcers distinguishes competition law from many
other fields. For instance, victims that suffer large-scale low-value damages are generally
consumers and they could equally benefit from collective devices in other consumer-oriented
fields. However, the existence of a competition-specific public enforcer enables the creation
of additional incentives for the initiation of collective ADR. This interplay between public and
private enforcement turns competition law into a valuable testing ground for collective ADR.
1.3 A toolbox of collective redress
Before starting the exploration of collective ADR as a tool of private enforcement, some basic
considerations must be provided in favour of collective proceedings. In essence, the lower
transaction costs of aggregated proceedings make them a necessary tool for the achievement
of efficient private enforcement outcomes in large-scale low-value damage scenarios. The
economic advantages of collective proceedings over individual proceedings are obvious in
relation to the lower individual costs borne by the victims. Savings can also be envisioned for
the defendants, who would not have to engage on several fronts and could benefit from the
certainty of once-and-for-all outcomes. Aggregation is also capable of lowering the broader
public costs resulting from the administration of justice. This would arguably not be the case
if collective proceedings lead to the introduction of claims which would otherwise not have
been brought at all. Yet, the multifaceted societal benefits resulting from the increased access
to justice and deterrence of wrongdoing could counterbalance the costs associated with the
additional claims arising from gap-filling areas.
Two categories of collective proceedings could be considered in order to reap the
aforementioned benefits: collective litigation and collective ADR. The former is a formal
process that requires initiation on behalf of the victims, whereas the latter is an informal,
Rodger, ‘The Consumer Rights Act 2015 and Collective Redress for Competition Law Infringements in the UK:
a Class Act?’ (2015) Journal of Antitrust Enforcement Vol. 3(2), 258, and B. Rodger, ‘Private Rnforcement of
Competition Law, the Hidden Story: Competition Litigation Settlements in the United Kingdom, 2000-2005’
(2008) European Competition Law Review Vol. 29(2), 96.
6
consensual process that requires interest from both sides. Both collective ADR and collective
redress litigation can take very different procedural forms, but the key characteristics of both
mechanisms are suitable for a comparative analysis. The pros and cons of both mechanisms
can be compared in the abstract, but for the purposes of this thesis, it is more important to
compare them in relation to achieving particular goals. In other words, the question pursued is
not whether we should choose between ADR and litigation, but rather, what should be their
relationship in different enforcement contexts. Ultimately, the success of collective private
enforcement depends on its effectiveness in delivering compensation and creating deterrence,
not on its particular mode of execution. It is proposed that any mechanism that is capable of
fulfilling these objectives should be given consideration.
Collective ADR and collective litigation share several structural characteristics that are
common to all collective proceedings. In particular, the effectiveness of collective
proceedings is largely based on whether the grouping of victims occurs on an opt-in or optout basis. The use of an opt-out model, whereby non-participating victims would be bound
without their explicit consent, raises important access to justice concerns, which must be
addressed. Moreover, all collective proceedings require some form of representation on behalf
of the victims, which raises principal-agent concerns. The representatives would also require
some funding in order to start the proceedings. While the costs of litigation can undoubtedly
be substantial, even the most informal negotiation-based forms of ADR are not costless.
Finally, in order to achieve sufficient finality, the international recognition and enforcement of
the collective outcome would have to be ensured. Although the shared characteristics of
collective proceedings raise common themes, differences will emerge in the way these
concerns are to be mitigated.
When comparing the two mechanisms, a general point of consensus is that the procedural
efficiencies derived from the speed, lower costs and flexibility of ADR would be unmatched
by litigation in most civil justice systems. Arbitration can be considered an exception in some
cases, since it has become more expensive in recent years. Although the resolution of mass
disputes would never be particularly fast, cheap or result in simple outcomes, it is likely that
collective ADR is able to outperform collective redress litigation due to the consent and
voluntary cooperation of the infringer. This points to the elephant in the room - the voluntary
nature of collective ADR. In other words, it is only capable of delivering redress if also the
infringing undertaking consents to it. This warrants the development of a regulatory
7
enforcement architecture, which creates incentives to induce the consent of the infringer. The
construction of this architecture will be one of the key objectives of the thesis.
There are additional ways in which collective ADR could work to the advantage of both the
infringers and victims better than collective litigation. Firstly, a negotiated outcome would
offer the certainty of obtaining actual compensation, which might not be the case with
litigation in case the infringer were to become insolvent as a result of lengthy litigation
proceedings. 8 Secondly, a ‘compromise’ in the context of collective ADR would not
necessarily represent a relinquishment of legitimate claims on both sides of the dispute, if one
considers that the value that the parties place on the particular aspects of the claim can be
different.9 Through negotiation, the parties would receive the opportunity for a smarter tradeoff that represents their real needs and interests. These ‘more precisely just’ trade-offs can
lead to unique remedial outcomes that would not be possible in courts.10 Thirdly, access to
evidence could be hindered by the uncooperativeness of the infringing undertakings in a
litigation setting. In a confidential ADR setting, they might be more likely to disclose certain
documents. Fourthly, collective ADR could also help overcome procedural obstacles that are
especially prevalent with competition law infringements. For example, there are potential
conflicts between victims that are direct and indirect purchasers, since the harm they suffered
is intertwined. When determining a common ‘class’ in collective litigation, some national
systems could prevent them from joining their efforts due to the lack of a common interest or
benefit.11 As long as the passing-on defence exists, conflicts between victims are capable of
undermining collective redress litigation.12 Collective ADR would be more flexible and allow
the parties to decide the scope of the coverage themselves. Another example regards the
serious difficulties with the quantification of harm due to the methodology and resources that
would have to be spent on obtaining the relevant data and expertise. For certain victims, it
would be impossible to gather the information required to make an accurate assessment of
specific and individual harm for the purposes of litigation.13 Again, the flexibility of collective
ADR would allow for custom approaches and approximations of harm.
8
H. Collins, Regulating Contracts (Oxford University Press 2002), p. 332.
C. Menkel-Meadow, ‘Whose Dispute Is it Anyway?: A Philosophical and Democratic Defense of Settlement
(In Some Cases)’ (1995) Georgia Law Review Vol. 83, 2663, p. 2673.
10
Menkel-Meadow 1995 (n 9), p. 2674-2675.
11
This scenario occurred in Emerald Supplies Ltd v British Airways Plc [2009] EWHC 741 Ch, [2010] CH 48.
12
L. Athanassiou, ‘Collective Redress and Competition Policy’ in A. Nuyts & N.E. Hatzimihail (eds), CrossBorder Class Actions: The European Way (Sellier European Law Publishers 2014), p. 153.
13
Athanassiou 2014 (n 12), p. 156.
9
8
Ultimately, if it is possible to achieve private enforcement outcomes for victims that would
otherwise not receive redress, all the while doing this in the most optimal manner with the
consent of all parties involved, then why would we not explore this further?
1.4 EU policy developments
Regrettably, collective ADR has not yet been served as a serious policy option for the private
enforcement of competition law nor within the broader collective redress debate. As a result,
there is a scarcity of literature assessing collective ADR and its interactions with the existing
redress and enforcement frameworks, especially with regards to EU competition law. In order
to place the forthcoming analysis of collective ADR into context, a brief overview of recent
EU-level developments on collective redress and private enforcement will be provided.
Legislative saga
The legislative initiatives on collective redress resulted from the parallel efforts by DG
Competition and DG Sanco. Although their policy actions were initially uncoordinated, they
were both guided by the enforcement gap for large-scale low-value damages in their
respective fields of EU law infringements. In 2005, DG Competition published a Green Paper
on Damages Actions for Breach of EU Antitrust Rules, which addressed the
‘underdevelopment of private enforcement’ and asked whether consumers and purchasers
with small claims should have a special collective litigation procedure.14 The 2008 White
Paper envisioned collective redress in the form of opt-in private actions and opt-out
representative actions for those representatives that constitute qualified entities.
15
Subsequently, the Commission proposed a Draft Directive that included these forms of
collective redress, but the Draft was removed from the agenda during the legislative stages.16
In the end, the Damages Directive 2014/104/EU on certain rules governing actions for
damages did not include any provisions for collective redress.17 Following a parallel track,
14
European Commission, ‘Green Paper on Damages Actions for Breach of the EC Antitrust Rules’ COM (2005)
672 final.
15
European Commission, ‘White Paper on Damages Actions for Breach of the EC Antitrust Rules’ COM (2008)
165 final.
16
L. Gorywoda, ‘The Emerging EU Legal Regime for Collective Redress: Institutional Dimension and Its Main
Features’ in A. Nuyts & N.E. Hatzimihail (eds), Cross-Border Class Actions: The European Way (Sellier
European Law Publishers 2014), p. 178.
17
Damages Directive (n 2), Recital 13: ‘This Directive should not require Member States to introduce collective
redress mechanisms for the enforcement of Articles 101 and 102 TFEU.’
9
DG Sanco published a Green Paper on Consumer Collective Redress in 2008, which outlined
the particular problems that consumers face and suggested collective litigation as a possible
solution.18 The 2009 Consultation Paper was a highly interesting follow-up, which, for the
first time, identified collective ADR as the redress option that was supported by the majority
of the respondents and suggested that an architecture could be developed with ‘sticks’ that
induce the initiation of ADR.19 As a stick, it not only envisioned the existence of collective
litigation, but also the empowerment of public authorities with skimming-off powers and
direct compensation orders. 20 As a non-legislative measure, it also suggested that an
exemplary collective ADR model could be developed and promoted on a voluntary selfregulatory basis, backed by independent monitoring. 21 Unfortunately, just like with DG
Competition, DG Sanco did not explore these options or propose any collective redress
initiatives after that. It is likely that the feedback received in response to their suggested
approaches had an impact on the Commission’s enthusiasm for collective redress and a
sector-specific solution. The inclusion of DG Justice in the debate led to a joint public
consultation by the three Directorates in 2011, aimed at finding a ‘coherent’ European
approach and signalling the end of developing sector-specific solutions in favour of a
horizontal collective redress measure.22 It also left out the further exploration of collective
ADR, as envisioned in DG Sanco’s 2009 Consultation Paper. This development, as it stands
today, has only resulted in a non-binding 2013 Commission Recommendation on common
principles for injunctive and compensatory collective redress. However, once the Commission
assesses the implementation efforts by the Member States within four years after the
publication of the Recommendation (26 July 2017 at the latest), legislative action could
follow.
Prevailing discourse
The current scene for collective redress is unsatisfactory not only because of its soft law
outcome, but also due to the positions that were adopted in the Recommendation, policy
documents and legal scholarship over the last decade. The Recommendation promotes an opt18
European Commission, ‘Green Paper on Consumer Collective Redress’ COM (2008) 794 final.
European Commission, ‘DG SANCO Consultation Paper’, Brussels, May 2009, p. 13-14.
20
European Commission 2009 (n 19), p. 17; the public enforcer powers were not deemed necessary in case an
EU-wide legislative instrument for collective redress would be adopted.
21
European Commission 2009 (n 19), p. 15.
22
European Commission, ‘Staff Working Document, Public Consultation: Towards a Coherent European
Approach to Collective Redress’ SEC (2011) 173 final.
19
10
in model, limits the standing of profit-making representative entities, prohibits contingency
fees and punitive damages, and instils the loser pays principle, which, in combination, are
unlikely to be financially incentivizing enough for victims with large-scale low-value damage.
The reason for these ‘safeguards’ is the fierce desire to avoid the US class action model,
which has been deemed to be a ‘toxic cocktail that should not be introduced in Europe.’23
Instead of adjusting particular aspects of the model to remove the possibilities for abusive
litigation, the Recommendation has arguably eliminated all financial incentives from the
procedure. Empirical evidence would unequivocally suggest that using an opt-in model for
large-scale low-value damages would be ineffective and result in a low participation rate.24 If
the same safeguards had been introduced in combination with an opt-out model, it would have
been more likely to achieve effective private enforcement for the weakest victims. Moreover,
an opt-out model is no longer foreign to the European legal tradition, as evidenced by the
numerous opt-out collective devices that have been adopted in recent years. 25 Yet, as it
currently stands, the adoption of a financially-incentivized European collective redress
litigation device appears to be politically unacceptable, which prompts the exploration of
alternative avenues.
In the competition law context, these problems have manifested in an additional way. The
tension between ‘deterrence’ or ‘compensation’ as the appropriate objective for EU private
enforcement has arguably contributed to the limited collective redress outcomes in the
Damages Directive. Emphasizing either objective would have an impact not only on the
policy choices regarding the relationship between public and private enforcement, but also
regarding the types of mechanisms that will be used for private enforcement. The US has
approached private enforcement from a deterrence perspective and therefore utilized
mechanisms, such as class actions, that are most likely to empower private parties to bring
claims. The EU has acknowledged both deterrence and compensation, but arguably, the
emphasis has been on the latter. With Courage and Manfredi, the Court of Justice established
a rights-based corrective justice narrative with the ‘right to full compensation’,26 which is also
23
European Commission, ‘Press Release of 27 November 2008, Green Paper on Consumer Collective Redress –
Questions and Answers’ MEMO/08/741, p. 4.
24
For example, see the comparative study R. Mulheron, ‘Reform of Collective Redress in England and Wales: A
Perspective of Need. A Research Paper for Submission to the Civil Justice Council of England and Wales’
(2008) Civil Justice Council of England and Wales.
25
Subject to limitations and sector-specific application, opt-out models are available in the UK, the Netherlands,
Denmark, Bulgaria, Belgium, Portugal and Germany.
26
Lianos, Davis & Nebbia 2015 (n 6), p. 22; Case C-453/99, Courage Ltd v Crehan [2001] ECR I-6297; Joined
cases C-295/04 to C-298/04 Manfredi and others [2006] ECR I-6619.
11
represented in Art. 3 of the Damages Directive. This perspective could result in favouring
enforcement mechanisms that, at least in theory, offer stronger protection for individual rights
at the expense of the potential deterrence benefits that could be achieved through opt-out
aggregation. Although the Court and the Commission have repeatedly acknowledged the
deterrence effects of private enforcement,27 its implications on collective redress mechanisms
have been limited. As discussed above, the development of a deterrence-oriented mechanism
was ultimately halted and the rights-based compensation narrative remains predominant.
1.5 Collective ADR in EU law
As evidenced by the policy developments on collective redress, there are no tailor-made
instruments focusing collective ADR on the EU level yet. However, individual forms of ADR
have received ample attention in the last decade, resulting in the establishment of minimum
safeguards that would also apply to collective ADR. Mediation has been a particular subject
of interest, as illustrated by the adoption of the 2008 Mediation Directive, which harmonised
several substantive and procedural elements.28 Consumers also benefit from several EU-level
safeguards. In particular, the Consumer ADR Directive was adopted in 2013, covering claims
arising from contractual obligations in sales or service contracts between consumers and
traders in both domestic and cross-border scenarios.29 It set minimum rules on transparency,
cost, duration, expertise, independence, impartiality and fairness of the ADR entities and
procedures, and granted monitoring tasks to the competent national authorities. Although its
scope is limited for the purposes of collective private enforcement of competition law, it
provides an overview of the minimum protections applicable in consumer ADR contexts, as it
largely mirrors the Unfair Terms Directive, the Brussels I Regulation and the relevant case
law. For instance, Art. 1 states that Member States are allowed to make ADR procedures
mandatory in B2C scenarios, as long as resorting to litigation after the procedure is still
possible. Art. 10 (1) ensures that pre-dispute agreements are unable to make the outcome of
the ADR procedure binding on the consumer. Finally, Art. 11 establishes that the outcome
reached cannot deprive the consumer of the mandatory protections from its Member State of
27
Ioannidou 2015 (n 4), p. 52-62.
Directive 2008/52/EC of the European Parliament and of the Council of 21 May 2008 on certain aspects of
mediation in civil and commercial matters, OJ L 136.
29
Directive 2013/11/EU of the European Parliament and of the Council of 21 May 2013 on alternative dispute
resolution for consumer disputes and amending Regulation (EC) No 2006/2004 and Directive 2009/22/EC
(Directive on consumer ADR), OJ L 165.
28
12
habitual residence. All of these safeguards must be taken into account when developing a
model of collective ADR.
The following sections will take a closer look at the two instruments that are most closely
connected to the introduction of a collective ADR mechanism for the private enforcement of
EU competition law - the Damages Directive and the Commission Recommendation on
Collective Redress.
Damages Directive
Despite often going unnoticed as one of the features of the Damages Directive,30 collective
ADR was acknowledged and encouraged in the different parts of the Directive. To begin with,
Recital 5 stipulates that litigation is just one facet of private enforcement next to ‘alternative
avenues’. The Directive highlights only one particular benefit of using ADR, which is,
somewhat surprisingly, not the speed or lower cost of the procedure. Rather, Recital 48
emphasizes the ‘certainty’ gained from what it refers to as ‘once-and-for-all’ solutions, which
should be encouraged by including as many victims and infringers in the consensual
settlement as possible. While such settlements include various efficiency gains, the emphasis
was placed on the broad scope of private enforcement that could be achieved through the use
of collective ADR. The different forms and effects of ADR that were prescribed in the
Directive will be assessed in more detail.
Consensual dispute resolution - Interestingly, the Directive does not use the term ‘alternative
dispute resolution’, but instead chooses to call it ‘consensual dispute resolution.’ Art. 2 (21)
defines consensual dispute resolution as ‘any mechanism enabling parties to reach the out-ofcourt resolution of a dispute concerning a claim for damages’ and the successful outcome of
the procedure is defined in Art. 2 (22) as a ‘consensual settlement.’ When comparing this
concept to the definition of ADR used in this thesis, they could readily be viewed as
synonyms. Perhaps the reason for using the term ‘consensual’ dispute resolution stems from
the fact that the Damages Directive primarily envisions a form of consensual negotiation,
which would lead to an agreement in the form of settlement terms. Yet, such a vision would
be unnecessarily limiting, although it may very well reflect the most common and efficient
ADR technique for achieving voluntary compensation. All of the other forms of ADR
30
Damages Directive (n 2).
13
identified in the ADR definition outlined in this thesis would also fall under the definition
provided by the Directive. There is no conflict because the other forms of ADR are also
consensual out-of-court processes that culminate with an agreement. The consensus is
expressed by agreeing to submit the dispute for consideration to a third party, such as an
arbitral tribunal, Ombudsman or ADR entity, and the decision that the third parties issue
would serve as the agreement or the settlement terms. The Directive confirms this
interpretation in Recital 48, which gives examples of consensual dispute resolution, which
cover not only negotiation-based ADR forms such as mediation and conciliation, but also
arbitration.
Incentives by the public enforcer - Returning to the broad definition of ADR, Recital 5 names
‘public enforcement decisions that give parties an incentive to provide compensation’ as an
example of an alternative private enforcement mechanism. The concept ‘public enforcement
decisions’ is only mentioned in this particular Recital and is not elaborated upon elsewhere in
the Directive using that terminology. The inclusion of this example acknowledges and
promotes the prospect of a public enforcer inducing voluntary compensation through the
provision of direct incentives. The only manifestation of this public and private enforcement
fusion can be found in Art. 18 (3), which allows competition authorities to take into account
successful ADR compensation as a ‘mitigating factor’ in setting the fine. This provision
strengthens the legitimacy of fine reduction initiatives that include collective ADR.
Suspensive effects - Two important suspensive effects of using collective ADR were clarified.
Firstly, Art. 18 (1) stipulates that the limitation period will be suspended for the duration of
ADR procedure. Secondly, Art. 18 (2) stipulates that ongoing litigation proceedings can be
suspended for up to two years if ADR procedures are initiated. Both the victim and the
infringer could benefit from the added time for building a stronger case for litigation or for
reaching a mutually satisfactory settlement.
Joint and several liability - In case of more than one infringing undertaking, the Directive
articulates the relationship between the infringers that decided to engage in collective ADR
and settle, and those that did not. In essence, Art. 19 (1) and (2) stipulate that the infringers
that decided to enter into a consensual settlement should no longer be held jointly and
severally liable along with the non-settling infringers. This is a coherent consequence of the
fact that the victim would no longer have any existing harm left vis-à-vis the settling infringer
14
as a result of the compensation received through the settlement. The Directive has thereby
attempted to increase the certainty felt by a settling infringer and encourage a favourable
setting for the use of collective ADR. However, this favourable setting is hindered by Art. 19
(3), which proceeds to give the Member States the discretion to derogate from that outcome in
case the non-settling infringers are ultimately unable to pay. Foremost, this situation can be
avoided altogether by explicitly excluding it in the settlement terms. If that was not done,
however, the goodwill of the setting infringer would be discredited by potentially forcing it to
compensate for the portion of the harm created by the insolvent co-infringers. As explained in
Recital 51, this turn of events would be justified by the need to guarantee the right to full
compensation. At the very least, Recital 52 urges national courts to ‘take into account’ the
amount of compensation already paid, so that the settling infringer would not have to pay the
full amounts remaining.
In sum, the Damages Directive has clearly promoted the use of collective ADR for private
enforcement. Important definitions, suspensive effects and joint and several liability
implications were outlined. Finally, the most interesting addition from the perspective of
integrating collective ADR into the broader competition law enforcement architecture - it
proposed the creation of incentives in the form of fine discounts for voluntary compensation
efforts.
Recommendation on Collective Redress
Although the bulk of the Recommendation 31 is focused on establishing principles for the
guidance of collective litigation procedures for injunctive and compensatory redress, it also
addresses and encourages the use of collective ADR. To begin with, paragraph 2 states that
the procedures designed in accordance with the principles of the Recommendation should be
‘fair, equitable, timely and not prohibitively expensive,’ which is also applicable to collective
ADR. In particular, the Recommendation highlights one benefit of using collective ADR efficiency. The different forms and effects of ADR that were prescribed will be assessed in
more detail.
Availability - Recital 16 stipulates that collective ADR should always be available in addition
to or as a voluntary step within the collective redress litigation proceedings. Furthermore,
31
Commission Recommendation (n 2).
15
paragraphs 25-26 specify that collective ADR can be initiated during the pre-litigation stage
as well as throughout the subsequent litigation stages prior to the issuing of a judgment.
Paragraph 25 appears to focus on negotiation-based ADR procedures, which aim to arrive at a
consensual out-of-court or court-based settlement. A reference was made to the Mediation
Directive, which harmonized the minimum standards applicable to mediation entities.
Whereas, paragraph 26 appears to address other ADR procedures (‘appropriate means of
collective ADR’) and specifies that their ‘use’, which is likely to mean their initiation, should
be based on consensus.
Suspensive effects - Paragraph 27 states that ongoing litigation proceedings will be suspended
for the duration of collective ADR proceedings. The Recommendation asserts that ADR
procedures can come to an end when one or both of the parties ‘expressly withdraw’ from it.
It is unclear how this would correspond to the functioning of collective ADR mechanisms,
which are initiated by consensual agreement, but are thereafter autonomous proceedings
resulting in a binding outcome issued by an ADR entity.
Judicial review - In case of a binding outcome in the form of a settlement, paragraph 28
stipulates that it should be subjected to court verification on ‘legality’ and that the
‘appropriate protection of interests and rights of all parties’ should be taken into account.
Again, it is unclear whether this form of judicial review is reserved exclusively for
settlements or whether ADR procedures that result in a binding outcome are also included.
Further, the substantive scope of the review was left ambiguous.32
In sum, the Recommendation on collective redress was brief yet explicit on the necessity of
ensuring the availability of collective ADR in addition to and throughout collective redress
litigation proceedings. However, the Recommendation was not entirely consistent or accurate
in its assessment of the procedures and outcomes resulting from different collective ADR
procedures. In particular, further distinctions need to be drawn between the procedures
resulting in a settlement and those resulting in a decision issued by an ADR entity.
Conclusion
32
European Law Institute, ‘Statement of the European Law Institute on Collective Redress and Competition
Damages Claims’, 2014, p. 42.
16
The current state of play regarding the development of effective and efficient EU collective
devices is unsatisfactory. The use of collective ADR as a private enforcement mechanism has
not yet been considered as a serious policy option. However, the introduction of collective
ADR into the private enforcement toolbox would be in perfect accordance with the pro-ADR
positions advanced in recent years, particularly in the Damages Directive and the
Recommendation on Collective Redress. Ultimately, it is submitted that collective ADR could
help close the enforcement gap for large-scale low-value damage if integrated into a
regulatory enforcement architecture that creates incentives for its voluntary use. The next
chapters will develop this hypothesis further.
17
18
Chapter 2: Incentives
Although collective ADR constitutes a distinct private enforcement device, the conceptual
analysis developed in this section requires looking at the broader context in which it
functions. In particular, the analytical framework will make use of the term ‘regulatory
enforcement architecture’, which refers to the sum of the public and private enforcement of
competition law. Put differently, it is suggested that the effective functioning of collective
ADR necessitates some degree of fusion between private and public enforcement, and that
coordination between the two can be mutually beneficial for the achievement of compensation
and deterrence outcomes. 33 With that in mind, the following questions will be explored
further: How could the voluntary use of collective ADR be triggered? What should be its
relationship with the other components of the architecture? The normative objective of the
analysis is to identify key elements of an architectural model that is capable of successfully
integrating collective ADR and thereby providing compensation to all victims, particularly to
those suffering large-scale low-value damage. In order to trigger the ‘voluntary and
consensual’ use of collective ADR, the architecture needs to provide incentives for two
categories of entities: (1) the infringing undertakings, and (2) the victims and their
representatives. After exploring these incentives, this section will conclude by sketching the
basic structure of a behaviourally-informed regulatory enforcement architecture.
2.1 Infringing undertakings
Designing an architecture that induces ‘voluntary compensation’ efforts through collective
ADR from undertakings that have infringed competition law, calls for critical inquiries into
behavioural research and regulatory enforcement theories. In order to shed some empirical
light on the behaviour of infringers, this section will cover findings from social psychology,
management studies and self-regulatory compliance programmes. As a word of caution,
behavioural research is often criticised for the notorious difficulty of drawing general
conclusions from complex, conditional and relational findings. 34 Despite the possible
epistemic, institutional and normative objections to utilizing behavioural findings in law and
33
It is accepted that the private enforcement of competition law is capable of achieving not just compensation
but also deterrence objectives; European Commission 2008a (n 15), p. 3.
34
Addressing the ‘nudge-scepticism’ of using behavioural research for law and regulation, see P. Cserne,
‘Making Sense of Nudge-Scepticism: The Challenges to EU Law’s Learning from Behavioural Sciences’ in A.
Alemanno & A.-L. Sibony (eds), Nudge and the Law: A European Perspective (Hart Publishing 2015).
19
regulation, certain data has found widespread recognition and will be discussed below.35 As a
final disclaimer, behavioural research will be explored in order to confirm the possibility of
infringers initiating collective ADR, which will be viewed as a form of compliance. The
effects of behavioural research on competition policy, substantive competition law and its
public enforcement - meaning compliance more broadly - are not explicitly assessed.
Undertakings with economic, social and normative motivations
As a starting point, undertakings will be viewed as multidimensional entities that operate in
complex contexts under the influence of laws and regulation.
36
These entities can
simultaneously be viewed as collections of individuals with different characteristics and
ambitions, tied together through organizational structures. Individual behaviour is impacted
by an organizational structure that incorporates the intersections of different interests,
behaviours and norms. Yet, the public enforcement of EU competition law is largely based on
a more one-dimensional view of undertakings, which rests upon the neoclassical
microeconomic assumption that they are unitary and rational utility-maximizing entities that
only comply with the law if it is their economic interest to do so. 37 In recent decades,
enforcement strategies from other disciplines have started to build upon the rationality
assumption and often employ a two-dimensional distinction between ‘good faith’ and ‘bad
faith’ undertakings. The good faith undertakings are viewed as having internalised moral
norms and are deemed willing to act upon them under certain circumstances, whereas bad
faith undertakings would only consider norms as external restraints and act exclusively on the
basis of economic self-interest. 38 It is implied that for rational, bad faith undertakings,
infringements would be conscious decisions to violate the law.39 If the rationality assumption
is a guiding principle of the enforcement strategy, then a conclusion that generally follows is
that enforcement design should be aimed at the bad faith undertakings, who require a
35
A. Alemanno & A.-L. Sibony (eds), Nudge and the Law: A European Perspective (Hart Publishing, 2015), p.
3.
36
C. Parker, The Open Corporation: Effective Self-Regulation and Democracy (Cambridge University Press
2002), p. 5.
37
A.R. Mele & P. Rawling (ed), The Oxford Handbook of Rationality (Oxford University Press 2004), p. 381384.
38
Y. Feldman & H.E. Smith, ‘Behavioral Equity’ (2014) Journal of Institutional and Theoretical Economics Vol.
170, 137, p. 138.
39
S.S. Simpson & M. Rorie, ‘Motivating Compliance: Economic and Material Motives for Compliance’ in C.
Parker & V.L. Nielsen (eds), Explaining Compliance: Business Responses to Regulation (Edward Elgar
Publishing 2011), p.60.
20
sufficient amount of deterrence in order to comply.40 Although this is a convenient paradigm,
empirical findings suggest that the distinction between the two categories might not be clearcut in all circumstances, since even good faith undertakings are capable of displaying
opportunistic behaviour. 41 This can be particularly pertinent if their goodwill is not
acknowledged and given an appropriate outlet. A fundamental conclusion that will be
elaborated upon later, is that focusing on the bad faith undertakings and using an exclusively
deterrence-oriented strategy would lead to a failure to make use of the untapped potential of
good faith undertakings and compliance-oriented avenues. It is clear that for undertakings that
exclusively require economic incentives to initiate collective ADR, the architecture must
simply make voluntary compensation an economically rational option. However, the main
focus of the subsequent discussions on behavioural research is on the exploration of noneconomic incentives and how to best make use of them.
Behavioural findings have put forth several considerations that explain why undertakings - as
collections of individuals with cognitive limitations - behave in a particular way. The
following five considerations will be deemed most relevant for explaining collective ADR,
when viewed as voluntary compensation or a form of compliance: (1) Individuals respond to
social influences, which means that they are influenced by what they perceive to be the
behaviour, knowledge and beliefs of others around them. 42 Therefore, when individuals
perceive that others want to comply, then they could be more likely to comply themselves.
However, various organizational pressures are capable of undermining such compliance
motivations, e.g. the encouragement of risk-taking to advance, the lack of personal
responsibility towards the outcome, the submission to group decisions in order to avoid
conflicts and the discouragement of whistleblowing in the name of loyalty to the
undertaking. 43 Some of these individual-organization tensions could be overcome by
connecting the individual will to comply with the profitability and economic rationality of the
undertaking. This connection is often exemplified by making a ‘business case’ for selfregulatory compliance programmes and corporate social responsibility. 44 (2) Individuals
40
I. Ayres & J. Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (Oxford University
Press 1992), p. 20.
41
C. Hodges, Law and Corporate Behaviour: Integrating Theories of Regulation, Enforcement, Compliance and
Ethics (Hart Publishing 2015), p. 30-31.
42
A. Alemanno & A.-L. Sibony (eds), Nudge and the Law: A European Perspective (Hart Publishing, 2015), p.
4.
43
Parker 2002 (n 36), p. 33-36.
44
E.C. Kurucz, B.A. Colbert & D. Wheeler, ‘The Business Case for Corporate Social Responsibility’ in A.
Crane (ed), The Oxford Handbook of Corporate Social Responsibility (Oxford University Press 2008), p. 85-92.
21
could be more likely to comply when they perceive the compliance objectives and processes
to be legitimate and procedurally fair. 45 This suggests that the promotion of a genuine
compliance culture could increase the individual motivation towards to comply. For instance,
successful self-regulatory compliance programmes generally have the following features:
managerial support and commitment, integration into decision-making procedures, designated
compliance personnel, public regulator pressures, private stakeholder pressures through civil
justice systems, and stakeholder involvement through public opinion. 46 (3) Individuals
experience inertia, meaning they tend to stick to the status quo and not work towards longterm change, even if it is in their best interest to do so.47 Therefore, if the status quo within an
organization is to comply, then individuals are more likely to accept it as a given and act
accordingly. (4) Individuals are influenced by the framing and presentation of information.
Ambiguity and flexibility about the process of compliance could increase ‘moral
deliberations’ on how to comply with the given rules and objectives. 48 In some contexts,
leaving discretion on how comply could create space for compliance motivations to emerge.
(5) Individuals have been found to have difficulties in the rational assessment of the
probabilities and risks involved with certain types of conduct. 49 This suggests that
infringements could sometimes be the result of poorly calculated misperceptions and
‘entrepreneurial delusion’, rather than perfectly calculated economic thinking.50 For instance,
decision-making is often based on intuition, information is taken into account selectively and
when the economic or personal stakes are higher, the likelihood of taking risks becomes
greater.51 Ultimately, it is submitted that a fuller understanding of the cognitive limitations of
individuals and organizations might improve the effectiveness of existing enforcement
strategies.
45
T.R. Tyler, ‘The Psychology of Self-Regulation: Normative Motivations for Compliance’ in C. Parker & V.L.
Nielsen (eds), Explaining Compliance: Business Responses to Regulation (Edward Elgar Publishing 2011), p.
84-85. This has been challenged by some research; for an ADR-based example regarding voluntary compliance
with Ombudsman decisions, see N. Creutzfeldt, ‘How Important is Procedural Justice for Consumer Dispute
Resolution? A Case Study of an Ombudsman Model for European Consumers’ (2014) Journal of Consumer
Policy Vol. 37, 527.
46
Parker 2002 (n 36), p. 50.
47
C.R. Sunstein, ‘Empirically Informed Regulation’ (2011) The University of Chicago Law Review Vol. 78(4),
1349, p. 1350-1351.
48
Feldman & Smith 2014 (n 38), p. 147.
49
Sunstein 2011 (n 47), p. 1358-1359.
50
Hodges 2015 (n 41), p. 23.
51
S. Bhattacharya & R. Van den Bergh, ‘The Contribution of Management Studies to Understanding Firm
Behaviour and Competition Law’ (2014) World Competition Vol. 37(4), 517, p. 521-522.
22
While these behavioural findings provide some insights into how to fine-tune the incentives
provided by the regulatory enforcement architecture, they also reveal that human behaviour
can be a complex variable, which is subject to a number of influences other than law and
regulation. Some undertakings will comply for the sake of complying, others can be
persuaded or threatened to do so; some undertakings are perfect profit-maximizing entities,
whereas others do not live up to that standard. Undertakings could also be motivated by a
combination of factors, which can change through time and depend on the context and
individuals involved. The best a regulatory enforcement architecture can do is to respond to
this plurality of behavioural considerations by providing an avenue towards compliance
outcomes for as many undertakings as possible. In doing so, deterrence should be viewed as a
necessary backup to ensure the success of the architecture, but compliance should be
facilitated as the first choice strategy for those that would be willing to comply if given the
right incentives. As will be elaborated later in the section, the integration of collective ADR
into the regulatory enforcement architecture would be illustrative of such a ‘complianceoriented’ strategy.
Case study: Lufthansa and the air cargo cartel
Unfortunately, there are no empirical studies about the particular behavioural processes that
have led to the initiation of collective ADR in a competition infringement scenario. However,
an illuminating example of such behaviour could be brought from the recent air cargo cartel,
which involved price-fixing behaviour on surcharges between some of the biggest airline
companies in the world, spanning between 1999-2006. 52 Lufthansa ‘discovered’ the cartel
shortly after creating an internal compliance programme and proceeded to approach all of the
competition authorities in relevant jurisdictions to report the cartel with a plea for leniency.
The public prosecution of the air cargo cartel resulted in more than 2,5 billion euros worth of
fines worldwide,53 with 799 million euros fined by the European Commission in 2010.54 The
public enforcement was followed by private enforcement efforts, including collective redress
52
The infringement decision was subsequently appealed and the General Court annulled most of the decisions in
December 2015; Cases T-9/11 Air Canada, T-28/11 Koninklijke Luchtvaart Maatschappij, T-36/11 Japan
Airlines, T-38/11 Cathay Pacific Airways, T-39/11 Cargolux Airlines International, T-40/11 Latam Airlines
Group and Others, T-43/11 Singapore Airlines and Others, T-46/11 Deutsche Lufthansa and Others, T-48/11
British Airways, T-56/11 SAS Cargo Group and Others, T-62/11 Air France-KLM, T-63/11 Société Air France
and T-67/11 Martinair Holland v Commission.
53
H. Bergman & D.D. Sokol, ‘The Air Cargo Cartel: Lessons for Compliance’ in Beaton-Wells & Tran (eds),
Anti-Cartel Enforcement In A Contemporary Age: Leniency Religion (Hart Publishing 2015), p. 302.
54
European Commission, ‘Antitrust: Commission Fines 11 Air Cargo Carriers €799 Million in Price Fixing
Cartel’ IP/10/1487, Brussels, 9 November 2010.
23
litigation in the form of class actions in the US. Lufthansa’s actions in the air cargo cartel are
illuminating not just for the role that their (1) self-regulatory compliance programme and (2)
leniency application played, but also for their (3) voluntary use of collective ADR to
compensate the victims even prior to the publication of the infringement decision. Thanks to
the research conducted by Bergman and Sokol,55 we have insights into the behind-the-scenes
development of these steps through to the interviews with Lufthansa’s in-house legal and
compliance counsel. The aim of this sub-section is not to summarize their work, but to point
out the key empirical findings that are of special relevance for understanding voluntary
behaviour in the context of this thesis.
Lufthansa is a large undertaking with diverse airline services and total annual revenues
exceeding 30 billion euros.56 Its Compliance Office was created in 2003 at the persistence of
the General Counsel von Ruckteschell and as a response to the compliance failure scandals in
the US and EU at the time. Compliance with competition law was deemed as the main
priority of the programme and was made binding as internal policy in 2004. Lufthansa made
genuine attempts to develop a compliance culture with regards to competition law by securing
the support of top executives, as well as by educating managers and employees through
lectures and meetings, both in person and online.57 According to Lufthansa, employees were
encouraged to speak up and report any concerns without suffering adverse consequences. Due
to the lack of a strictly hierarchical corporate structure, employees were able to cooperate and
communicate freely between different units. Lufthansa also stressed the importance of
integrating lawyers into its business operations, so that their presence would become a natural
part of the decision-making processes. Shortly after these compliance activities had been
initiated, employees reported to the Compliance Office that price-fixing activities with other
airline companies had been going on for years and that such ‘negotiations’ were thought to be
legal or even required by law.58 When the findings were presented to the executive board, it
allegedly took them only 10 minutes of deliberation to authorize bringing the cartel to the
attention of competition authorities and seeking leniency. Lufthansa’s compliance and legal
teams started collaborating with external legal counsel and embarked upon full cooperation
with competition authorities in 2006 - an expensive process that lasted for three years, but
ultimately, Lufthansa received full immunity in all jurisdictions. Shortly after the
55
Bergman & Sokol 2015 (n 53), p. 301-314.
Lufthansa, ‘Annual Report 2014’, Frankfurt, March 2015.
57
Bergman & Sokol 2015 (n 53), p. 308-309.
58
Bergman & Sokol 2015 (n 53), p. 310.
56
24
investigations became public, Lufthansa became the target of 20 class action claims in the US
and one in Canada. According to Lufthansa, their decision to engage into collective ADR and
seek an out-of-court settlement was the result of: (1) a cost-benefit analysis, and (2)
reputational considerations.59 Despite paying out 85 million euros in settlements,60 some of
which were considered to be ‘overstated figures’ by the claimants, Lufthansa believes that by
voluntarily settling the claims even prior to the rendering of the infringement decision, they
not only saved money by avoiding class action proceedings, but also protected their
reputation in the airline industry.61
In sum, it is clear that voluntary behaviour in the form of whistleblowing to receive leniency
is considered to be an indispensable element of the EU competition enforcement architecture.
Leniency as a form of compliance fits well into the deterrence-oriented enforcement strategy
that public enforcement employs. The use of leniency is substantiated by the rationality
assumption, whereby under certain circumstances, betraying the cartel is assumed to become
economically more beneficial than continuing the participation in the collusive behaviour.62
Yet, other forms of voluntary behaviour such as self-regulatory compliance programmes or
voluntary compensation through collective ADR are not expressly promoted or rewarded.
This could be due to the inability to explain such compliance behaviour using a deterrenceoriented rationale. However, as evidenced by the experience of Lufthansa, the inclusion of
undertakings in the prevention, detection and mitigation of competition law violations should
not disregarded, since as a calculated element of a regulatory enforcement architecture it
could prove to be a source of untapped potential. This thesis focuses on its potential for
private enforcement, particularly through the encouragement of collective ADR and other
voluntary compensation efforts.
2.2 Victims and their representatives
Obtaining the consent of the infringing undertaking could be deemed as the biggest
impediment to the initiation of collective ADR, which is why the regulatory enforcement
architecture is predominantly oriented towards the infringers. Once the infringers are offering
to consider voluntary compensation, victims and their representatives would be likely to
59
Bergman & Sokol 2015 (n 53), p. 312.
Lufthansa, ‘Annual Report 2006’, Cologne, March 2007.
61
Bergman & Sokol 2015 (n 53), p. 312.
62
Bhattacharya & Van den Bergh 2014 (n 51), p. 517.
60
25
explore that opportunity. Next to the increasingly secure prospects of receiving compensation,
the victims and their representatives could also be incentivized the informal and non-judicial
nature of collective ADR proceedings, which generates speed, lower cost, confidentiality and
procedural flexibility. Next to these ‘inherent incentives’, there are two particular issues that
must be addressed for the successful functioning of collective ADR within the regulatory
enforcement architecture.
Victims - With collective ADR, the victims would constitute as ‘non-participating parties’ to
the proceedings, since they would ultimately be represented by a third entity. The direct
incentives that can be provided for the victims mainly occur at the stage, where they must
either opt-in or opt-out of the collective proceedings. In case the victims suffered large-scale
low-value damage, individual activity levels would be very low due to the low value of the
claim and the incentives that can be provided to spur any form of victim participation would
thereby be limited as well. The choice between an opt-in or an opt-out model is therefore
critical for the success of collective proceedings, for both the ADR and litigation varieties.
Using opt-in as the default regime has been empirically proven to lead to lower participation
rates, especially with low-value claims, whereby the victims might often not be aware of their
claim at all. 63 To illustrate, the only opt-in collective competition damages claim brought
under the UK’s pre-reform Competition Act 1998 was the 2007 JJB Sports case, which
achieved compensation for approximately 0.1% of the victims affected.64 In comparison, the
Dutch opt-out settlement regime produced a participation rate of approximately 97% of the
victims affected in the 2007 Dexia securities case. While the exact figures can vary, it is clear
that any type of activity levels - whether to opt-in or opt-out - are very low with low-value
claims, so if the aim of the enforcement architecture is to ensure that as many victims as
possible receive compensation, then an opt-out model should be preferred. As discussed in
chapter 1, the EU-level policy discussions on collective redress were largely focused on
avoiding opt-out models for a number of reasons, including the desire to avoid a US-style
litigious culture and to safeguard individual procedural rights. This was reflected in the 2013
Commission Recommendation on Collective Redress, which explicitly promoted an opt-in
63
Z. Juska, ‘Obstacles in European Competition Law Enforcement: A Potential Solution From Collective
Redress’ (2014) European Journal of Legal Studies Vol.7(1), 125, p. 143. For an overview on the advantages
and disadvantages of both models, see Mulheron 2008 (n 24).
64
UK Department for Business Innovation & Skills, 'Private Actions in Competition Law: A Consultation on
Options for Reform’, April 2012, p. 11; The Consumer Association v JJB Sports Plc (2007) CAT Case No.
1078/7/9/07.
26
model as the default option, only allowing opt-out in exceptional cases.65 However, in light of
the considerations above, this approach should be reconsidered. The possibility of an opt-out
model for collective litigation would increase the threat of litigation and thereby induce the
use of collective ADR within the regulatory enforcement architecture. Likewise, the
possibility of an opt-out model for collective ADR would increase its attractiveness for the
infringers, who could thereby achieve international ‘once-and-for-all’ solutions and limit their
exposure to multiple claims for an extended period of time. In sum, although the exact scope
of the coverage for collective ADR outcomes it is ultimately up to the parties to consensually
decide, the architecture should at least facilitate the possibility of an opt-out application to
ensure the overall success of the system for large-scale low-value claims.
Representatives - The direct incentives that the architecture could create for representative
entities regard their funding. Although collective ADR is likely to involve fewer expenses
than collective litigation, even the most informal negotiation-based forms of ADR are not
costless. Once the infringer’s consent has been induced by the architecture, then the offer to
start collective ADR proceedings or to receive a complete redress scheme can be incentivizing
enough for the representative entities or their funders due to the high certainty of receiving
their share of the compensation. However, particularly in case the threat of collective
litigation needs to be established in order for collective ADR to be induced in the first place,
additional financial incentives for representative entities would be needed. There are at least
six categories of funding sources that could be relevant for collective redress scenarios: (1)
contingency fees, (2) legal aid, (3) individual contributions, (4) insurance, (5) special funds,
(6) third party funding. When considering the merits of these funding options, it should be
kept in mind that the 'loser pays’ principle, 66 which is applied in most Member States, 67
already creates major disincentives to start an action, which is why a combination of adequate
funding mechanisms is imperative. Contingency fees allow the representative to make
representation fees contingent on a successful outcome, which could incentivize law firms in
particular; however, only a few Member States allow them and the Commission 2013
65
Commission Recommendation (n 2), Recital 21, exceptions allowed in case such actions would be ‘duly
justified by reasons of sound administration of justice’.
66
The ‘loser pays’ principle stipulates that the losing party would have to cover the legal costs of the opposing
party.
67
Juska 2014 (n 63), p. 133.
27
Recommendation on Collective Redress explicitly prohibited such practices.68 National legal
aid is often very limited and unlikely to cover cross-border situations. 69 Individual
contributions are likely to be insignificant in case the victims suffered large-scale low-value
damage. Likewise, insurance for legal expenses is likely to be restricted when it comes to
collective proceedings. Special funds could be a great solution, whether European or national,
if pooled with the help of donations from successful claimants, confiscated profits, injunctive
orders, competition law violation fines, cy-près damages or crowdfunding. 70 Finally, third
party funding could be the most viable option, whereby the representative entity would be
funded by another entity that would receive a portion of the final compensation amount.71
Such funding could also be provided on a contingency basis, however, the 2013 Commission
Recommendation on Collective Redress prohibited this form of contingency as well. 72
Overall, the particular funding arrangements that are deemed most appropriate for collective
proceedings could be left for the Member States to determine. When making such policy
decisions, it should be considered that each funding arrangement could be individually
scrutinized during an ex post fairness review of the final outcomes of collective ADR.
Ultimately, representative entities should have sufficient incentives to represent victims in
collective ADR proceedings, as long as Member States do not create significant restrictions
for third party funding and legal counsel fees.
2.3 Designing a regulatory enforcement architecture
In order to develop a well-informed architecture that takes behavioural intricacies and
multilayered incentives into account, this section will start exploring enforcement theories
that can help guide the manner in which collective ADR is integrated into the architecture.
Enforcement theories aim to uncover effective and efficient strategies that would ensure the
fulfillment of the regulatory objectives. However, these theories often present idealised
68
Commission Recommendation (n 2), Recital 31, exceptions allowed in case Member States ‘provide for
appropriate national regulation of those fees in collective redress cases, taking into account in particular the right
to full compensation of the members of the claimant party’
69
European Law Institute 2014 (n 32), p. 33.
70
European Law Institute 2014 (n 32), p. 34.
71
R. Mulheron, ‘Third Party Funding and Class Actions Reform’ (2015) Law Quarterly Review Vol. 131(Apr),
291, p. 294.
72
Commission Recommendation (n 2), Recital 32, the only exception: ‘unless that funding arrangement is
regulated by a public authority to ensure the interests of the parties’.
28
hypotheticals that are unlikely to be found in the real world in their purest prescribed form.73
At one end of the spectrum, there exist several variations of the command-and-control theory,
which is still the primary approach used by competition law public enforcers. These strategies
are based on the idea that the prescribed standards of conduct will be enforced using coercion,
deterrence or ‘punishment’ in the form of criminal or civil sanctions. At the other end of the
spectrum, there are incentive-based theories, which do not rely on coercion, but rather attempt
to ‘persuade’ the regulatees into compliance.74 In between these two extremes lie enforcement
theories that can be considered a more accurate representation of what an optimal regulatory
enforcement mix might look like in practice. They entail a pluralistic, responsive, reflexive
and smart approach that provides an avenue for different punish-and-persuade synergy
scenarios of inducing compliance. These intermediate enforcement theories are deemed most
relevant for guiding the use of collective ADR and will be explored further.
As the seminal work that sparked a mass of research in the field of regulatory enforcement,
Ayres and Braithwaite’s ‘responsive regulation’ theory takes into account the plurality of
motivations that different undertakings have for complying with rules and concludes that a
one-size-fits-all strategy would be ineffective. Instead, the theory prescribes that the regulator
should respond to the behaviour of undertakings in a tit-for-tat manner, with the starting point
always being voluntary compliance. The starting point of voluntary compliance is important
as not to punish the undertakings that are willing to comply, in addition to the fact that, from
the regulator’s perspective, persuasion would generally be cheaper than punishment.75 If the
undertaking does not comply voluntarily, then the regulator can escalate their response in an
enforcement pyramid that leads up to the strongest sanction, which could be, for instance, a
criminal penalty.76 It is essential that a regulator possesses a ‘big stick’ penalty, however, this
stick would ideally never have to be used for more than a light ‘tap on the shoulder’,77 since
its very existence would be enough to induce voluntary compliance. 78 The theory of
responsive regulation has spanned a family of different theories that are based on the same
core idea - responsiveness - but which move beyond some of the limits set by the enforcement
73
N. Gunningham, ‘Enforcement and Compliance Strategies’ in R. Baldwin, M. Cave & M. Lodge (eds), The
Oxford Handbook of Regulation (Oxford University Press 2010), p. 122.
74
N. Gunningham, ‘Strategizing Compliance and Enforcement: Responsive Regulation and Beyond’ in C.
Parker & V.L. Nielsen (eds), Explaining Compliance: Business Responses to Regulation (Edward Elgar
Publishing 2011), p. 199.
75
Ayres & Braithwaite 1992 (n 40), p. 19.
76
Ayres & Braithwaite 1992 (n 40), p. 35-39.
77
C. Hodges, The Reform of Class and Representative Actions in European Legal Systems: A New Framework
for Collective Redress in Europe (Hart Publishing 2008), p. 212.
78
Ayres & Braithwaite 1992 (n 40), p. 40-41.
29
pyramid.79 The follow-ups have also highlighted some criticisms about the appropriateness of
a rigid gradual escalation in serious risk and single interaction situations. 80 For instance,
Gunningham and Grabosky’s ‘smart regulation’ theory builds upon responsive regulation by
adding new surrogate regulators and regulatory tools to the mix. It includes commercial and
non-commercial actors and third parties, who can deploy different regulatory methods to
affect behaviour.81 The theory does not limit itself to escalating enforcement action within a
single enforcement pyramid and allows for escalation interactions between the different sides
of a three-dimensional enforcement pyramid by different actors.82 While the focal point in
each of these responsive theories can differ, they share common principles that can prove to
be helpful even when applied in broader enforcement contexts.
Responsive private enforcement: carrots and sticks
A straightforward application of the enforcement theories described above to ‘private
enforcement’ would not be entirely sound. The controversy stems from the fact that private
enforcement is not a form of ‘enforcement’ as it is traditionally understood.83 In particular,
enforcement by private actors through the civil justice system does not always fit easily
within the enforcement theories that are built upon the archetype of a public regulator.
Moreover, the objective of the civil justice system is not governance in the direct sense as it is
with public regulation. However, modern enforcement theories commonly recognise that a
wide range of actors and tools could be utilised to achieve the desired outcomes. This
connects to what some authors have identified as a broader ‘transformation of enforcement’,
which also regards the functions and methods traditionally deployed for public and private
79
Many enforcement theories in the responsive regulation family do not concentrate on the compliance and
deterrence dichotomy in a hierarchical setting, but mainly highlight additional aspects of enforcement strategy.
For instance, Baldwin and Black’s ‘really responsive regulation’ theory focuses on a holistic performance
assessment by extending the rationale of responsiveness to additional aspects of the broader regulatory
enforcement architecture, such as the undertakings’ internal frameworks, broader institutional frameworks,
interactions between different regulatory strategies, performance of the regulatory architecture, and finally,
towards any changes to the aforementioned elements; R. Baldwin & J. Black, ‘Really Responsive Regulation’
(2008) Modern Law Review Vol. 71(1), 59, p. 61. Another example regards ‘risk- or problem-based regulation’
theories, which focus on the objectives and targeting of enforcement rather than the particulars of the regulatory
response; R. Baldwin, M. Cave & M. Lodge, Understanding Regulation: Theory, Strategy, and Practice (Oxford
University Press 2012), p. 268.
80
Baldwin & Black 2008 (n 79), p. 62-64.
81
Baldwin, Cave & Lodge 2012 (n 79), p. 266.
82
Gunningham 2011 (n 74), p. 200.
83
D. Ashton & D. Henry, Competition Damages Actions in the EU: Law and Practice (Edward Elgar Publishing
2013), p. 3, fn. 9.
30
enforcement.
84
When abandoning a binary distinction between public and private
enforcement, a more functional hybrid approach can emerge.85 If the responsive theories were
applied in a functional manner to a regulatory enforcement architecture integrating collective
ADR, then the following principles would emerge: (1) An exclusively deterrence-based
approach is not effective as a first choice strategy. Behavioural findings would even suggest
that sticking with a confrontational deterrence strategy could lead to the development of
defensiveness and intentional non-cooperation on behalf of the infringing undertakings.86 The
cooperative and consensual nature of collective ADR as the compliance-oriented first choice
strategy would offer an outlet for the distinct economic, social and normative motivations that
undertakings possess. Only in situations where collective ADR was unable to result in
compensation, should the enforcement escalate to collective litigation, which serve a
deterrence-oriented function. (2) Enforcement results could be reached more effectively and
efficiently by granting a degree of freedom to the regulatees for figuring out how to meet the
regulatory goals, while making full use of their resources and knowledge.87 Collective ADR
could provide this flexibility and directly involve the infringing undertakings in the
dispensing of compensation, especially when compared to litigation alternatives.
Furthermore, using the information and resources of the infringing undertakings to dispense
redress could be more cost-effective than using the resources of the court system and thereby,
the resources of the victims and other non-victim taxpayers. (3) The regulatory enforcement
strategy should not undermine the desirable characteristics of the social practice that it is
ultimately regulating.88 This would be particularly prevalent in case the pro-competition and
pro-consumer goodwill that manifests in the form of voluntary compensation and other
compliance efforts would not be taken into account by the architecture. The rewarding of
collective ADR could take various direct and indirect forms, which will be outlined in more
detail below.
84
See Micklitz & Wechsler 2016 (n 1).
C. Hodges & N. Creutzfeldt, 'Transformations in Public and Private Enforcement' in H.-W. Micklitz & A.
Wechsler (eds), The Transformation of Enforcement: European Economic Law in Global Perspective (Hart
Publishing 2016), p. 115. For a discussion on the hybrid approach in competition law, see C. Hodges, ‘European
Competition Enforcement Policy: Integrating Restitution and Behaviour Control’ (2011) World Competition
Vol. 34(3), 383.
86
Gunningham 2011 (n 74), p. 201.
87
J. Stapleton, ‘Regulating Torts’ in C. Parker and others (eds), Regulating Law (Oxford University Press 2004),
p. 128.
88
H. Collins, ‘Regulating Contract Law’ in C. Parker and others (eds), Regulating Law (Oxford University Press
2004), p. 25.
85
31
Building upon the principles above, a responsive strategy would suggest that compensation
becomes more likely if infringing undertakings were provided a range of private enforcement
avenues to follow, while presenting collective ADR as the first choice. These avenues could
be offered by various surrogate regulators, who do not necessarily have to be public enforcers.
A typical collective ADR avenue would result in a consensual agreement between the
infringing undertaking and the victims as surrogate regulators, which would fulfill the
regulatory objective of compensation. This process could be aided by various incentives
provided by public enforcers, courts, victims, third parties, institutional and procedural
characteristics, which would take the form of a compliance-oriented carrot or a deterrenceoriented stick. Although both sets of incentives are aimed at inducing voluntary
compensation, the difference in their presentation could result in distinct behavioural effects
on undertakings, who possess diverse motivations and capacities to comply.
Carrots - A direct carrot incentive could be a fine reduction that is issued by the public
enforcer in case the collective ADR took place before or in the course of the investigation
procedure. The fine reduction could result from considering compensation as a mitigating
factor in the fining guidelines or as a result of following a more regulated collective ADR
avenue prescribed by the public enforcer. Such incentives would be less attractive for those
infringers that received full immunity from the fines through leniency, unless considerations
would be given to an amendment to the leniency policy that adds compensation as a condition
for full immunity. Next to the public enforcer incentives, collective ADR also holds inherent
carrot incentives derived from its procedural characteristics, which would have to be
facilitated by the architecture. As an informal and non-judicial device, it offers speed, lower
cost, confidentiality and procedural flexibility. As a voluntary device, it harbours distinct
reputational benefits that could be used by infringers, whose business models are sensitive to
public perception. As a consensual device, it allows the infringers to display cooperative traits
that can be used to remedy pre-existing business or consumer relationships. Depending on the
procedural rules applicable to collective ADR regarding its opt-in or opt-out application, and
the recognition and enforcement of its outcomes, it could serve as the device for achieving
once-and-for-all international settlements, which allows the infringers to avoid multiple
dispute resolution processes and minimize their global exposure to litigation.
Sticks - A direct stick incentive could be the threat of collective litigation. Legislators and
courts could provide additional incentives by taking the prior use of collective ADR into
32
account; for instance, when deciding whether a claim should proceed on an opt-in or opt-out
basis, or when determining the distribution of the costs of litigation. Moreover, certain
procedural characteristics of collective ADR that were highlighted as potential carrot
incentives for some infringers, could serve as stick incentives for others in case collective
litigation operates with the same characteristics. In particular, the broad coverage of victims
through opt-out application and guaranteed recognition and enforcement, could make the
threat of collective litigation more intense. Even if the alternative is to enter into collective
ADR proceedings with the same characteristics, the informal and flexible arena could be the
‘lesser evil’ for many infringers. Next to litigation, the refusal to try collective ADR, when
proposed by the victims, could serve as an aggravating factor in the fining guidelines, which
would be analogous to the fine reductions offered by the public enforcer. Since collective
ADR is ultimately a voluntary device, then the aggravating aspect of their refusal would have
to be supported by additional circumstances, such as the deliberate misleading of the victim
representatives with the purpose of increasing their financial expenditure or weakening their
potential litigation claim.
As evident from the discussions above, there are limits to what the biggest deterrence stick
can be within the boundaries of ‘private enforcement’. Unless even greater fusion between
private and public enforcement is sought, then the public enforcer’s powers will be limited to
influencing the consensual compensation process by the private actors. Further fusion
between public and private could result in a stronger stick; for example, the public enforcer
could directly order the payment of compensation themselves or have the opportunity to
represent the victims in collective litigation proceedings. However, the architecture
envisioned in this thesis will limit itself to private enforcement with mere inducements from
public enforcers, since proposing further measures would warrant a deeper assessment of the
roles of public and private enforcement of competition law, which is not the aim of the
present inquiry.
89
While public enforcers are given an important role in achieving
compensation within the regulatory enforcement architecture envisioned in this thesis, it will
not go as far as to suggest that ensuring compensation should be the direct task of public
enforcers.
89
For a discussion on competition authorities dispensing redress, see Hodges 2011 (n 85) and Ioannidou 2015 (n
4), p. 171.
33
In sum, the attraction of a responsive enforcement strategy is derived from the plurality of
actors, incentives, behavioural considerations and approaches involved. If behavioural
findings suggest that the affected parties would use collective ADR under certain
circumstances, then the architecture should set up the relevant incentives in a way the induces
such outcomes. Ultimately, when executed effectively, using collective ADR as a tool of
private enforcement would not only result in compensation for the victims, but it would
achieve it in the most optimal manner with the consent of all parties involved.
34
Chapter 3: Safeguards
A wholesome understanding collective ADR as a tool of private enforcement should not only
focus on its role within a regulatory enforcement architecture, but also on its implications for
the civil justice right to the restoration of harm. This section will tread more traditional
themes associated with ADR and collective redress, and address the following questions:
What should be the procedural characteristics of a collective ADR device in order to achieve
effective and efficient outcomes? How to ensure fairness guarantees within informal out-ofcourt collective proceedings that feature non-participating victims? The analysis subjects
collective ADR to the scrutiny of access to justice considerations in an attempt to establish
additional elements that should inform its use within the architecture envisioned. When
addressing the criticisms, risks and rights-capabilities of collective ADR, comparisons will be
drawn against its formal adjudicative counterpart, collective litigation.
3.1 When, how and under what circumstances
Few would fundamentally dispute the potential contributions of ADR to increasing access to
justice, seeing as the last decades have moved the question of whether the use of ADR is
appropriate from ‘for or against’ to ‘when, how and under what circumstances.’ 90 In
particular, concerns can arise when ADR is applied collectively in mass harm cases and in
areas that fall within the public policy domain; both of which would be present when
proposing the use of collective ADR for the private enforcement of competition law. On a
conceptual level, the critique of collective ADR in this context is largely based on the private
nature of the redress mechanism and the inherent value that is bestowed upon judicial
determination. Despite their formal independence, courts are an extension of state power and
while they have the capacity to resolve disputes, that is not their only task.91 The courts are
considered to be in charge of dispensing justice and taking into account the values of the civil
justice system when issuing a judgment. With collective ADR, whether an adjudicative ADR
entity issues a decision or the parties reach a mutually pleasing compromise, such outcomes
could be viewed as being far from achieving justice due to the lack of a public dimension. In
line with this limiting view, ADR entities are seen as mere rent-a-judge entities that are
90
Menkel-Meadow 1995 (n 9), p. 2665.
S. Roberts & M. Palmer, Dispute Processes: ADR and the Primary Forms of Decision-Making (Cambridge
University Press 2005), p. 221.
91
35
considered suitable to deal with issues of fact, but not with issues of law.92 Consequently, in
sensitive areas that touch upon critical societal values, proponents of this view would argue
that it is more important to have justice through courts than to have peace through ADR.93 In
essence, these criticisms imply that collective ADR is exclusively based on private interests
(often only the individual interests of the stronger party) and ultimately void of values that
only a public dimension could provide. This argument is likely to constitute an
oversimplification that places such private processes into a legal and societal vacuum.
Proponents of ADR submit that consensual dispute resolution mechanisms are based on
several values that correspond to those promoted by the civil justice system; such as
participation, harmony, respect, empathy, consensus, privacy, community and equity.94 While
the courts receive their legitimacy from their connection to the state, collective ADR
procedures are legitimized by the consent of the individual parties. Some even view ADR as a
social movement or a value system, which can act as a democratic assertion of individual
justice or as an ideology that introduces a new dimension to civil justice.95 Moreover, the
advancement of private interests with ADR does not exclude the parallel advancement of
public interests, which can occur through increased deterrence. The criticisms addressed
above also disregard the broader processes of formalisation and informalisation that
increasingly blur the line between public and private. The informalisation is evidenced by the
development and widespread acceptance of ADR mechanisms, the inclusion of ADR in court
proceedings and tasking the judges with case settlement objectives; whereas formalisation
constitutes the increased regulation and juridification of ADR mechanisms, through which
informal ADR processes are integrated into a formal framework to meet due process
requirements and obtain court-like qualities. 96
The extent to which the concerns about the loss of a ‘public dimension’ are relevant when
discussing the private enforcement of competition law, can vary. It will be argued that
criticisms of such nature have a limited impact in the scenarios that involve the use of
collective ADR under the architecture envisioned in this thesis, for the following reasons:
92
A. S. Kim, ‘Rent-a-judges and the Cost of Selling Justice’ (1994) Duke Law Journal Vol. 44, 166, p. 166.
O.M. Fiss, ‘Against Settlement’ (1984) Yale Law Journal Vol. 93, 1073, p. 1085.
94
Menkel-Meadow 1995 (n 9), p. 2670.
95
M. Palmer, ‘Formalisation of Alternative Dispute Resolution Processes: Some Socio-Legal Thoughts’ in
Zekoll, Bälz & Amelung (eds), Formalisation and Flexibilisation in Dispute Resolution (Brill 2014), p. 17;
Menkel-Meadow 1995 (n 9), p. 2676.
96
Palmer 2014 (n 95), p. 23, 42-44.
93
36
(1) It should be considered that using collective ADR - a voluntary device - is likely to occur
predominantly in the course of on-going public investigations or as a follow-on claim after
the infringement decision. The follow-on nature of the claims will effectively mitigate several
concerns. It is less likely that undertakings would voluntarily agree to take proactive
compensatory action in stand-alone claims without the existence of any investigations or
infringement decisions, unless the substantive and procedural rules of a legal system are
sufficiently favourable to materialize a serious threat of stand-alone collective litigation.
(2) The use of collective ADR would occur directly in the ‘shadow of the law and the
courthouse’, meaning the dispute would be assessed in reference to the rules of competition
law, tort law and civil procedure, as well as considering the infringement decision, the
likelihood of impending private damages litigation and its possible outcome. The consensual
resolution of the dispute would therefore not be unprincipled or detached from the law, as the
critics of ADR often fear; it would be directly dependant on the relevant legal variables.
Consequently, if there is uncertainty about the content or application of the relevant rules,
then it is likely to be reflected in the collective ADR process.
(3) The precedential and educational value of claims, which would be lost if collective ADR
were to be used instead of litigation, should not be overstated. From the perspective of
clarifying substantive competition law, follow-on claims could essentially be deemed
useless.97 The potential contribution of such claims would be limited to clarifying procedural
tort elements, such as establishing harm and causation. The contribution of stand-alone claims
would be more significant, however, the proposed addition of collective ADR as first choice
would never exclude the use of litigation. This means that the less straightforward cases that
contain procedurally or substantively uncertain claims are less likely to be settled voluntarily
through ADR to begin with and would be presented in courts, where the law could be
clarified. Furthermore, such legal issues could equally be clarified through legislative action
or mass torts from other disciplines.
(4) Finally, not all disputes trigger the need for the public and judicial clarification of laws
and societal values. The development of collective ADR should not be hindered by simply
classifying all mass torts as requiring public oversight in the form of litigation due to the
97
W. Wils, ‘The Relationship between Public Antitrust Enforcement and Private Actions for Damages’ (2009)
World Competition Vol. 32(1), 3, p. 5.
37
amount of parties involved. With follow-on actions in particular, concerns regarding the
‘public interest’ and the ‘need for the avoidance of abuse’ have largely been taken into
account already when the initial decision by the public authority was rendered.98
In sum, the scenarios that involve the use of collective ADR under the regulatory enforcement
architecture envisioned in this thesis, do not pose serious concerns that negate the use of
private dispute resolution mechanisms. However, collective ADR is not without its
challenges, which nevertheless warrants the development of public oversight.
Consenting to collective ADR
The key concern with collective ADR is not its non-judicial and private nature, but the
existence of genuine consent to the proceedings and to its binding outcome. As a starting
point, it is clear that any inequalities of power that would exist with collective litigation
would also exist with collective ADR. The infringing undertakings would continue to possess
economic and information advantages in informal processes. It is also true that the
confidentiality of the proceedings combined with the relaxation of procedural requirements
would not always work in the favour of the weaker party.99 The basic rhetoric of consensus
assumes a level of equality between the parties, often envisioning ‘a quarrel between two
neighbours’, which could be best solved by coming to a ‘compromise.’ 100 This idealized
vision of private autonomy does not represent diffuse large-scale low-value damages, which is
why an aggregation of private claims would be necessary to balance the scales. However,
since consent can be considered a cornerstone of ADR, it would be important to determine
whether genuine consent can be retained with collective ADR if several non-participating
victims are involved. Post-dispute multi-party proceedings that only cover the participating
parties and result in a binding agreement are not per se problematic or likely to encounter
significant difficulties with the recognition and enforcement of the outcome. Yet, the
attraction of collective proceedings in mass harm situations lies in its ability to deliver onceand-for-all solutions with broad coverage of the relevant victims. Achieving such outcomes
could necessitate the use of an opt-out model, whereby the outcome would be made binding
98
The same argument is made in Commission Recommendation (n 2), Recital 22 - “Moreover, in the case of
collective actions following a decision by a public authority (follow-on actions), the public interest and the need
to avoid abuse can be presumed to have been taken into account already by the public authority as regards the
finding of a violation of Union law.”
99
Roberts & Palmer 2005 (n 91), p. 54.
100
Fiss 1984 (n 93), p. 1076.
38
on victims that never explicitly agreed to be included in the proceedings. If an opt-out
application is considered to be problematic even in the context of litigation, then it would
require even more careful scrutiny in the context of ADR. Furthermore, introducing collective
ADR could produce additional consent-related scenarios that need to be considered. For
instance, it makes a difference whether the agreement to engage in collective ADR occurs
before or after the dispute has arisen. Moreover, in case the victims include consumers,
concerns can arise regarding their ability to consent to the proceedings in the first place, given
the inequalities of power with the infringing undertakings. These concerns have a direct
impact on determining the level of public oversight that would be required to ensure due
process and fairness guarantees. The following analysis will feature a mix of positive and
normative evaluations by first outlining the state of play in the EU and then assessing its
implications for the regulatory enforcement architecture.
No consent - The most straightforward problem with consent would arise if it were to be
eliminated altogether by making the ADR proceedings mandatory by law. With consumers as
victims, the Court of Justice established in Alassini that making ADR proceedings mandatory
would not violate the consumer’s right to effective judicial protection, as long as it does not
force the consumer to be bound by the outcome and create disproportionate costs or delay.101
Therefore, the key problem lies with the finality of the ADR proceedings, meaning that
mandatory ADR that results in a binding outcome (i.e. arbitration) would be prohibited.
However, non-binding forms of ADR would be allowed because if the parties fail to come to a
consensual agreement, then they could always resort to litigation. Thus, in B2C situations,
Member States have the discretion to make non-binding forms of ADR mandatory. In B2B
situations, businesses are subject to different treatment under national laws and are unlikely to
enjoy similar protections. In any case, the mandatory nature of collective ADR proceedings
has not yet been explicitly addressed, but the complexities of mass harm are likely to trigger
more careful consideration under national laws. Even so, the utility of making non-binding
collective ADR mandatory would be rather low in the first place, since it would not lead to
the certainty of once-and-for-all outcomes if the victims could choose not to be bound by the
outcome after all. Ultimately, the aim of the regulatory enforcement architecture is to create
different avenues for private enforcement and promote collective ADR as the first choice
through facilitation and inducements, not on a mandatory basis.
101
Joined cases C-317/08-C-320/08 Alassini and others v Telecom Italia [2010] ECR I-2213.
39
Pre-dispute consent - Additional problems with consent can arise in case the infringing
undertakings and their victims have concluded a pre-dispute agreement that stipulates the use
of ADR for subsequent disputes related to or arising from the contract. Although many
victims will not have any contractual connection to the infringing undertakings, it is likely
that some of the potential parties to the collective proceedings will. Such an agreement can
take the form of a standardized clause in a contract, which would generally be accepted by the
victim due to their lack of knowledge or bargaining power. Again, the key problem regards
the validity of pre-dispute consent in case of binding ADR proceedings. In B2C situations, the
Unfair Terms Directive would render such clauses that have not been individually negotiated
as unfair in case they ‘exclude or hinder the consumer’s right to take legal action.’102 Member
States have interpreted the consequences of the unfairness differently - either by making the
clauses automatically void or simply avoidable under certain conditions - but most predispute binding ADR clauses with consumers would be prohibited. 103 In B2B situations,
businesses are subject to different treatment under national laws and are unlikely to enjoy
similar protections. However, some Member States are also considering the adoption of
consumer-like unfair terms protections for SME-s. 104 Again, the pre-dispute consent to
collective ADR proceedings has not been explicitly addressed yet, but the limitations with
consumers would apply in any case and weaken the desirability of the mechanism for the
infringers. Some Member States have introduced special rules for opt-in multi-party
arbitration, e.g. consumer arbitration in Spain and shareholder arbitration in Germany.105 The
particular finality of pre-dispute arbitration clauses that aim to bind victims in an opt-out
manner warrants further analysis. The US ‘class arbitration’ experience can prove to be an
illuminating case study and will be discussed further below.
Post-dispute explicit consent - Obtaining the post-dispute consent of every victim in an
explicit manner is the least problematic scenario. Ideally, all relevant victims would be
sufficiently informed and join collective ADR proceedings voluntarily after the infringement
has taken place, which would amount to an opt-in model. As discussed in chapter 1, an opt-in
model was considered the most ‘European’ and politically acceptable approach throughout the
collective redress policy developments. However, despite the obvious merits of an explicit
102
Council Directive 93/13/EEC on unfair terms in consumer contracts, Art. 3 (1) & Annex (q).
I. Bantekas, An Introduction to International Arbitration (Cambridge University Press 2015), p. 259-260.
104
UK Department for Business Innovation & Skill, ‘Protection of Small Businesses When Purchasing Goods
and Services: Call for Evidence’, March 2015.
105
S.I. Strong, Class, Mass, and Collective Arbitration in National and International Law (Oxford University
Press 2013), p. 19-20.
103
40
consent model, it must be reconciled with the reality that the participation rates with opt-in
proceedings are very low in case of low-value damages due to economic, procedural, social
and cognitive reasons. 106 Moreover, the limited coverage of victims would undermine the
attractiveness collective ADR for the infringing undertakings and therefore be unlikely to
solve the private enforcement gap. Although the choice regarding the extent of collective
ADR’s coverage is ultimately up to the parties to decide, an opt-in model would not serve as
the most potent utilisation of the device.
Post-dispute implicit consent - Obtaining the post-dispute consent of non-participating
victims in an implicit manner by including them in the proceedings unless they explicitly opt
out, is more problematic from an individualistic procedural justice perspective. However, as
discussed above, an opt-out is model capable of producing superior participation rates and is
ultimately necessary to ensure the effective private enforcement of low-value damages. The
individual due process and fairness concerns could be mitigated by a fairness review, which
will be addressed later in this section. Ultimately, an opt-out model in combination with
adequate safeguards should be facilitated to secure compensation for as many victims as
possible and to increase the attractiveness of voluntary compensation for the infringing
undertakings. The ‘facilitation’ of an opt-out model within the architecture means providing
the possibility to receive approval and recognition of the opt-out coverage during the fairness
review and for its subsequent cross-border enforcement.
Case study: class arbitration
A special category of collective ADR could illustrate an additional consent-centred dynamic.
The preceding discussions have primarily focused on obtaining the consent of the victims,
with the assumption that the infringers would be interested in collective ADR as a result of
the incentives provided by the architecture. However, the infringers could also use arbitral
clauses (or other contractual ADR agreements) in an abusive way to limit their exposure to
collective proceedings. As a starting point, the characteristics of arbitration prescribe that
every potential participant of collective arbitration must have an arbitral agreement with the
infringing undertakings. If the victims do not have an arbitral agreement with the infringing
106
Discussed in chapter 2 regarding the incentives for victims and their representatives; for further analysis, see
Mulheron 2008 (n 24).
41
undertakings, then it will not be possible to use arbitration in any manner.107 If the victims do
have an individual arbitral agreement, then it would be possible to join their claims for
collective arbitration. The joinder might involve a preliminary certification phase whereby the
arbitral tribunal determines the suitability of the relevant victims.108 If the joinder occurs on
an opt-in basis, then it would simply be a multi-party form of arbitration, which does not raise
concerns per se. However, if the joinder occurs on an opt-out basis, then a similar set of
concerns that were discussed in the previous paragraph would arise. Opt-out collective
arbitration is commonly referred to as ‘class arbitration’ due to its US origin. In particular, the
US experience has highlighted three scenarios of consenting to class arbitration:109
(1) The agreement explicitly prescribes class arbitration - Although rare in practice, when
given the right incentives, class arbitration would ideally be used in this manner. In the US,
the existence of such an arbitration agreement would allow to bind non-participating parties
on an opt-out basis. The EU does not yet feature class arbitration regimes.
(2) The agreement is silent on class arbitration - The US Supreme Court dealt with silence on
two occasions. Initially, in Green Tree v Bazzle (2003) the possibility of class arbitration in
case of silence was left for the arbitral tribunal to decide. 110 This would have been in
accordance with the Kompetenz-Kompetenz principle that allows arbitrators to assess their
own competence to arbitrate. However, the situation changed with Stolt-Nielsen (2010),
which established that ‘arbitration is fundamentally based on consent’ and silence could
therefore not constitute an acceptance of class arbitration.111 It is likely that a similar position
would be adopted in most Member States, given the lack of class arbitration regulation and
the importance of consent in regular arbitral proceedings. In case the infringer is not
cooperative, this would work against the victims and prevent them from joining their claims.
If silence were to be interpreted as implicit consent, then questions could arise about whether
the ‘nature of arbitration’ is affected. An affirmative interpretation would be more victim-
107
It could be possible to convert a settlement into an arbitral award, but not in case the settlement included
victims without pre-existing arbitral agreements on an opt-out basis; S.I. Strong, ‘From Class to Collective: The
De-Americanization of Class Arbitration’ (2010) Arbitration International Vol. 26(4), 493, p. 506.
108
American Arbitration Association, ‘Supplementary Rules for Class Arbitration’ (AAA 2003) and Judicial
Arbitration and Mediation Services, ‘Class Action Procedures’ (JAMS 2009) are most frequently used as
specialized guidance; Strong 2013 (n 105), p. 44.
109
P. Billiet (ed), Class Arbitration in the European Union (Maklu 2013), p. 12; Strong 2013 (n 105), p. 171172.
110
Green Tree Financial Corp. v. Bazzle, 539 U.S. 444 (2003).
111
Stolt-Nielsen S. A. v. AnimalFeeds Int’l Corp., 559 U.S. 662 (2010).
42
friendly, especially since the unexpected class arbitration could have easily been avoided by
the infringing undertakings by drafting more explicit arbitral clauses.
(3) The agreement explicitly prohibits class arbitration with a class waiver - Such waivers
compel the victim to individual arbitration and explicitly exclude class arbitration. Following
AT&T v Concepcion (2011), waivers were found to be ‘conscionable’ and thereby permitted
even in standardised contracts with consumers.112 In the antitrust case American Express v
Italian Colors Restaurant (2013), waivers were not found to impede the right to ‘effective
vindication’ even if resorting to individual arbitration would make it too expensive for any
action to be taken at all. 113 Allowing undertakings to effectively eliminate any threat of
collective proceedings - both ADR and litigation - through individual arbitration clauses, is
very concerning and likely to magnify existing inequalities of power. In this regard, the US
experience can serve as an example of the choices to avoid. Considering the high level of
consumer protection in the EU, pre-dispute non-negotiated class waivers that exclude
litigation would be prohibited. However, with businesses as victims, the situation would vary
per Member State.
In sum, the existence of genuine consent to collective ADR is a multi-layered issue that
requires an assessment of all consent-related scenarios that could emerge. As collective ADR
has not yet been regulated on the EU level, there are many gaps in the law. To date, the EU
has introduced minimum safeguards that protect consumers from certain abuses, but in case
the victims are businesses, their protection would be subject to different treatment in the
Member States. Ultimately, particular attention must be paid to an opt-out application of
collective ADR, whereby many non-participating victims would be included on the basis of
implicit consent.
3.2 Balancing incentives and safeguards
As evidenced by the preceding discussions, collective ADR is exposed to certain risks and
tradeoffs, especially in its most potent opt-out form. This warrants the development of public
oversight in order to ensure due process and fairness guarantees. It is imperative, however, to
develop a balanced approach towards regulating collective ADR’s weaknesses as not to
112
113
AT&T Mobility LLC v. Concepcion, 563 U.S. 333 (2011).
American Express Co. v. Italian Colors Restaurant, 570 U.S. ___ (2013).
43
diminish its strengths. In particular, as a private dispute resolution mechanism, collective
ADR should not receive the exact same treatment as collective litigation. While some
tradeoffs are inevitable with any form of collective proceedings, it will be argued that the key
concerns with collective ADR can be mitigated by the architecture. The following sections
will first dissect the different ‘outcomes’ that collective ADR proceedings can produce and
then suggest how the fairness concerns could be addressed.
Once-and-for-all: recognition and enforcement
Voluntary engagement in collective ADR would not be as desirable for the undertakings if
they were uncertainties about the cross-border recognition and enforcement of the outcome of
the proceedings. Achieving ‘once-and-for-all’ solutions that cover as many victims as
possible, has been highlighted as one of the most desirable features of collective ADR.114
Unfortunately, the recognition and enforcement of collective ADR outcomes remains a
problematic issue, since in some contexts their res judicata effect is not viewed as a benefit,
but as a risk. Since the EU is still relatively unfamiliar with collective ADR, the situation is
worsened by the presence of gaps in the private international law regime. For a streamlined
analysis, a distinction will be made between the different possible outcomes of collective
ADR proceedings.
ADR entity decisions - The decisions issued by ADR entities are accepted as binding on the
basis of a contractual agreement between the parties. In order to bind victims in an opt-out
manner, court approval would be needed to secure sufficient legal certainty. While this is a
common route for negotiated mass settlements, ADR entities have barely any experience in
this regard. Generally, compliance with ADR entity decisions is only ensured within national
borders with specific enforcement measures in the national laws.115 Adding the requirement of
judicial approval for every decision is likely to undermine the autonomy, functioning and
attraction of ADR entities, unless their role is reduced to settlement facilitators.
114
See the Damages Directive (n 2), Recital 48: “Achieving a ‘once-and-for-all’ settlement for defendants is
desirable in order to reduce uncertainty for infringers and injured parties. Therefore, infringers and injured
parties should be encouraged to agree on compensating for the harm caused by a competition law infringement
through consensual dispute resolution mechanisms, such as out-of-court settlements (including those where a
judge can declare a settlement binding), arbitration, mediation or conciliation. Such consensual dispute
resolution should cover as many injured parties and infringers as legally possible. The provisions in this
Directive on consensual dispute resolution are therefore meant to facilitate the use of such mechanisms and
increase their effectiveness.” [emphasis added]
115
For an overview on the coverage of consumer ADR entities in Europe, see C. Hodges, I. Benöhr & N.
Creutzfeldt-Banda (eds), Consumer ADR in Europe (Hart Publishing 2012), p. 392.
44
Settlements - As with the decisions of ADR entities, settlements are accepted as binding on
the basis of a contractual agreement between the parties and in order to bind victims in an optout manner, court approval would have to be secured. The Brussels I Regulation Recast
clarified the status of settlements in Art. 2 (b), which declared both litigation settlements and
out-of-court settlements that received court approval, as ‘court settlements.’ Therefore, the
court that approves such settlements needs to assert jurisdiction in accordance with Brussels I
and if the settlement is successfully approved, Art. 58-59 stipulate that it must be enforced in
all Member States. The only exception arises in case the settlement would be manifestly
contrary to public policy. Public policy concerns can arise with settlements that cover opt-out
victims due to a lack explicit consent. The requirements regarding the consent of such victims
have not been clarified in the Brussels I Regulation or by the Court of Justice. In this regard,
the Court has referred to the choice of forum provision (Art. 25) of Brussels I, which can be
satisfied with implicit consent.116 However, the question of whether ‘not using the opportunity
to opt-out’ is considered as implicit consent has not been clarified and Member States can
develop different approaches, which is capable of threatening the effectiveness of opt-out
settlements. The boldest cross-border opt-out jurisdiction assertions occur in the Netherlands
and will be assessed further in chapter 4.
Arbitral awards - Arbitration is excluded from the scope of the Brussels I Regulation by Art. 2
(d) as the recognition and enforcement of arbitral awards is governed by the New York
Convention. Multifaceted competition damages claims would be covered due to the broad
interpretation of ‘commercial disputes’.117 However, the intricacies of collective arbitration,
especially opt-out class arbitration, are largely ungoverned by the existing regime.
Nevertheless, Art. V of the Convention includes several grounds for refusal, which could
become relevant for class arbitration. Firstly, if the applicable law does not allow for class
arbitration, then the underlying arbitral agreement could be deemed invalid.118 Secondly, the
agreement could also be invalid if the subject matter is not considered arbitrable in that state.
This could occur in case the arbitral award attempts to bind EU consumers in a mandatory
116
A. Stadler, ‘Mass Damages in Europe – Allocation of Jurisdiction – Cross-border Multidistrict Litigation’ in
W.H. van Boom & G. Wagner (eds), Mass Torts in Europe: Cases and Reflections (Walter de Gruyter 2014), p.
214.
117
Strong 2010 (n 107), p. 524-525.
118
Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York, 1958), Art. V
(1)(d).
45
manner.119 Thirdly, if the class arbitration proceeding did not include sufficient due process
for the non-participating parties, such as notification and an opportunity to participate, then
the award could be refused.120 The sufficiency of the procedural safeguards will generally be
interpreted in accordance with the law of the seat.121 Fourthly, the award could be refused if
the opt-out models would breach the public policy of the enforcement state.122 However, the
relevant standard is that of ‘international public policy’, which requires serious substantive or
procedural concerns and is rarely a successful refusal basis in practice. 123 All in all, these
grounds for refusal are capable of threatening the effectiveness of class arbitration.
In sum, the cross-border recognition and enforcement of ADR outcomes that attempt to bind
non-participating victims in an opt-out manner, is a problematic issue regardless of the type of
collective ADR procedure chosen. Regardless of the opting model used, if the aim of the EU’s
private international law regime is facilitate collective proceedings, then the existing
legislation could benefit from numerous procedural clarifications.124 Although this thesis will
not go into depth about the possible private international law reform proposals in this context,
it is acknowledged that improving the enforcement regime is imperative for the effective
functioning of the regulatory enforcement architecture and could be achieved by introducing
special provisions on collective redress in the Brussels I Regulation. The following subsection will delve further into mitigating the issues that underpin the resistance to collective
ADR’s finality and broad coverage.
Fairness review
Collective proceedings with non-participating parties produce specific tensions that must be
addressed by the legal system in which it operates or by the system to which it is submitted
for recognition and enforcement. As a normative starting point, the elements that require
balancing are not the traditional principles of procedural justice and the efficiencies of
collective ADR, but rather, two different notions of procedural justice. A distinction could be
made between the ‘individual fairness’ of binding victims out-of-court with representatives,
especially in an opt-out manner, and the ‘collective fairness’ that includes the victims,
119
New York Convention (n 118), Art. V (2)(a).
New York Convention (n 118), Art. V (1)(b).
121
Z.S. Tang, Electronic Consumer Contracts in The Conflict of Laws (2nd edn, Hart Publishing 2015), p. 327.
122
New York Convention (n 118), Art. V (2)(b).
123
Billiet 2013 (n 109), p. 27; Strong 2010 (n 107), p. 541-547.
124
Stadler 2014 (n 116), p. 201.
120
46
infringers and the society as a whole. Collective fairness encompasses the access to justice
gains achieved for victims that would not have been compensated otherwise, the public
benefits of deterrence, the removal of illicit gains,125 the promotion of consensual dispute
resolution and the once-and-for-all coverage benefits for the infringers. Collective fairness is
greater than the sum of its individual fairness parts.126 Put differently, using collective ADR
with large-scale low-value claims produces positive externalities that go beyond the
vindication of separate or joined individual rights. Instead of resorting to per se unfairness on
the basis of individualistic procedural justice, the broader implications of collective ADR
should be scrutinized while taking into account the relevant regulatory objectives and the
limitations of the existing enforcement frameworks. This balancing exercise adopts a broader
approach to access to justice, which is not limited to legal formalism and a single goal of
individual compensation. 127 Access to justice requires a contextual understanding of the
needs, solutions and impacts of the legal response, which takes into account the cultural,
economic and psychological realities of individual and organizational behaviour.
128
Ultimately, collective ADR provides more than just procedural efficiencies; it can advance
notions of collective fairness for all affected parties.
While some tradeoffs between individual and collective fairness are inevitable with any form
of collective proceedings, it is submitted that the the key access to justice concerns with
collective ADR can be mitigated by a fairness review. In order to preserve the autonomous
character of out-of-court collective ADR, the review and approval of the outcome should take
place after the proceedings have concluded. For stand-alone and follow-on claims, the review
can take place during the enforcement stage by the courts; for the follow-on claims that are
settled in the framework of ongoing investigations, the review could take place concurrently
with the public investigation and involve the public enforcer as a reviewer. In the latter
scenario, the involvement of neutral third party reviewers could be envisioned.129 Additional
court entanglement at an earlier stage would not only threaten the procedural efficiencies of
collective ADR, but also undermine the internalization of the due process and fairness
125
R. Fentiman, ‘Recognition, Enforcement and Collective Judgments’ in A. Nuyts & N.E. Hatzimihail (eds),
Cross-Border Class Actions: The European Way (Sellier European Law Publishers 2014), p. 107.
126
M. Shapiro, ‘Class Actions: The Class as Party and Client’ (1998) Notre Dame Law Review Vol. 73(4), 913,
p. 933; Ioannidou 2015 (n 4), p. 112.
127
Ioannidou 2015 (n 4), p. 115-117.
128
M. Cappelletti, ‘Alternative Dispute Resolution Processes Within the Framework of the World Wide Access
to Justice Movement’ (1993) Modern Law Review Vol. 56(3), 282, p. 283.
129
For instance, the ‘voluntary redress schemes’ under the UK architecture assign a significant review role for
independent experts, as will be discussed later in the thesis.
47
requirements in the dispute resolution process. 130 An ex post fairness review would put
pressure on the parties and, if applicable, on the relevant ADR entities, to meet these
requirements in a way that best accommodates the consensual nature of the process. In other
words, with ex post fairness review, the parties would have to work harder to present a
wholesome package that gives due regard to collective interests, whereas with ex ante
preliminary review, they would be partly relieved of this task by the court-determined class
certification. The following non-comprehensive benchmarks are a compilation of common
elements found in several collective litigation models and, depending on how they would be
applied in the national systems, have the capacity to represent a fair balance for the ex post
fairness review.
(1) Compensation - The review should not determine whether an even better outcome could
be reached, but whether the outcome that has been agreed upon is adequate. 131 It should
identify unequivocal signs of unfairness, conflicts of interest and a lack of reasoned
explanations with regards to the compensation amount agreed upon. 132 In particular, the
fairness of the compensation allocation between different categories of victims (i.e. direct and
indirect purchasers) and the compensation for the representative entity should be reviewed. It
should also consider that a negotiated outcome would offer the certainty of obtaining actual
compensation, which might not be the case with litigation in case the infringer were to
become insolvent. Moreover, the broadness of the coverage (i.e. choosing to include indirect
purchasers) could have warranted compromises in other aspects. All in all, a reasonable
review of the compensation achieved is warranted due to the existence of a large number of
non-participating parties to collective agreements.
(2) Notification and participation - The notice should provide minimum substantive and
procedural information about the dispute and the chosen dispute resolution procedure, the
parties involved, the representatives, the binding nature of the outcome and any possibilities
for direct participation. The victims that intend to opt-in or opt-out would have to receive
guidance as to the reasonable steps that they would have to take for that end. In case
confidentiality is insisted upon, the notice should specify the terms under which the recipient
130
C.J. Buckner, ‘Due Process in Class Arbitration’ (2006) Florida Law Review Vol. 58, 185, p. 238.
E.g. the US class action and class arbitration rules prescribe a ‘fair, reasonable and adequate’ standard,
although exact state-level interpretations diverge; see Federal Rules of Procedure Rule 23(e), AAA Rule 8,
JAMS Rule 6(a)(2); Buckner 2006 (n 130), p. 200.
132
J.R. Macey & G.P. Miller, ‘Judicial Review of Class Action Settlements’ (2009) Journal of Legal Analysis
Vol. 1(1), 167, p. 183-185.
131
48
may discuss the matters relating to the dispute. The specific manner in which the notification
requirement could be satisfied, can be left for the Member States to decide, but a ‘reasonable’
standard would be advisable in case the identification of victims is problematic.133
(3) Representation - The review should primarily identify any conflicts of interest between the
representative entity and the victims. The exact standards can be left for the Member States to
determine, especially in case they deem additional protections to be necessary for the opt-out
application. For instance, conflicts would be unlikely to emerge in case the entity is dedicated
to the protection of certain categories of victims, like consumers. The funding of
representative entities should be addressed under the general review of the compensatory
outcome.
(4) Plans for unclaimed compensation - Leftover funds are likely to occur with any low-value
claims due to the low motivation for taking action.134 This would be the case regardless of
whether an opt-in or opt-out model was used, although they are more likely with the latter. In
such cases, the development of cy-près solutions should be encouraged, especially if the
identification of victims and the distribution of compensation would be problematic.135 Cyprès would put the unclaimed funds in their ‘next best use’, for example by investing in a
victim- or infringement-related cause. In choosing the best cy-près outlet, emphasis should be
placed on contributing to the relevant collective interest, especially in order to justify an optout application. If a common collective interest is difficult to identify, then investing in access
to justice funds for future low-value claims could be considered. Simply reverting the
unclaimed compensation back to the infringing undertaking would be less likely to serve such
interests, unless it is coupled with a comprehensive compliance-oriented proposal. The
responsive and flexible nature of cy-près makes it a good fit for the regulatory enforcement
architecture.
Even with these benchmarks in mind, assessing the fairness of collective ADR outcomes
could undoubtedly be challenging for the reviewer, especially considering the presentation of
133
E.g. the US class action and class arbitration rules prescribe ‘reasonable efforts’ for developing the ‘best
notice practicable under the circumstances’; see Federal Rules of Procedure Rule 23(c)(2)(b), AAA Rule 6(a),
JAMS Rule 4.
134
As discussed above in relation to opt-in and opt-out models; for further analysis, see Mulheron 2008 (n 24).
135
R. Mulheron, ‘Cy-Près Damages Distributions in England: A New Era for Consumer Redress’ (2009)
European Business Law Review Vol. 20(2), 307, p. 329-331.
49
a non-adversarial ‘unified front’ and the difficulty of reaching a consensual solution.136 With
competition damages in particular, there could be contention regarding the distribution of
compensation between the different categories of victims and the approximations of harm
underlying the compensatory amounts. However, the fairness review ultimately provides the
opportunity to balance all of these considerations, while giving due regard to the advanced
stage of the dispute resolution process. In sum, in order to ensure the effective and efficient
functioning of collective ADR, the regulatory enforcement architecture needs to include
reasonable fairness safeguards for non-participating victims and an adequate private
international law regime for securing sufficient finality.
136
For a discussion on the challenges of approving cartel damages settlements in Canada, see R. Mulheron, 'A
Spotlight on the Settlement Criteria Under the United Kingdom's New Competition Class Action' (2016) Civil
Justice Quarterly Vol. 35(1), 14, p. 23-27.
50
Chapter 4: A comparative analysis of collective ADR in the Netherlands and the UK
A comparative analysis could illustrate how collective ADR works in practice and point
towards important considerations for the development of a European collective ADR model.
The Netherlands and the United Kingdom were chosen as the subjects of the comparative
study since, unlike most Member States, they both feature a national regulatory enforcement
architecture that integrates collective ADR. In order to develop a better understanding of
collective ADR as a tool of private enforcement, this chapter will be divided into two parts.
The first part provides a detailed overview of both national architectures, including an
assessment of the toolbox of collective redress proceedings that are available, as well as
insights into the capacities of the public enforcer to affect the provision of voluntary
compensation. Building on that information, the second part develops a comparative analysis
focusing on the key issues that determine the functioning and potential success of a collective
ADR model within both national systems.
4.1 The Netherlands
4.1.1 Toolbox of collective proceedings
The Netherlands does not have a competition-specific private enforcement regime featuring
collective ADR. However, the collective devices within the existing cross-sectoral regime
would also be applicable to competition damages claims. In particular, this section will
provide an overview of two devices: (1) opt-out collective litigation, and (2) opt-out collective
settlements. Next to these devices, the Netherlands features different collective joinder
procedures that bundle individually brought claims. Although they will not be included in this
section, these procedures bear a mention, since they have been used to obtain competition
damages.137 However, in case of large-scale low-value claims, the existence of the joinder
device would not create significant incentives for the voluntary initiation of collective ADR,
since the victims are unlikely to bring individual claims in the first place.
137
J.S. Kortmann & M.C. Buiten, ‘Private Enforcement and Collective Redress in European Competition Law’
(2015) FIDE 2016 Questionnaire, p. 20. For example, special purpose vehicles were created by the Cartel
Damages Claims group for the candle wax cartel, by the East-West Debt group for the elevator and escalator
cartels, and by several groups for the air cargo cartel.
51
Collective litigation
Less known than its settlement counterpart, the Netherlands has featured an opt-out collective
litigation device since 1994, housed in Art. 3:305a of the Dutch Civil Code. However, it is of
limited use for compensation claims as the collective proceeding can result in every remedy
except for compensatory damages. 138 The claims can only be brought by non-profit
representative entities that have a clear connection to the collective interest of the victims they
represent (i.e. in their articles of association). Despite lacking a compensatory remedy,
representative entities have sought injunctions and declaratory judgments with the aim of
setting the scene for subsequent individual claims. For instance, the biggest consumer
association in the Netherlands, the Consumer Union (Consumentenbond), has successfully
achieved declaratory judgments that resulted in follow-on damages claims. 139 The device
could also be used to induce settlements, since collective ADR is promoted as a pre-litigation
step in Art. 3:305a(2), which stipulates that the representative entity could only have a right to
action if their attempts at negotiation with the defendant were not successful. However, due to
the lack of a compensatory remedy and the difficulty of bringing individual follow-on claims,
the collective litigation device is ultimately unable to create any significant incentives for the
voluntary initiation of collective ADR for victims of large-scale low-value damage. In 2014,
the Dutch Minister of Security and Justice devised a preliminary draft bill with amendments
that would remove the prohibition on compensatory damages.140 The draft was accompanied
by a public consultation, but no legislative action has been taken yet.
Collective settlements
The Netherlands also features an opt-out collective settlement device that was meant to fill the
compensation gap left by collective litigation. The Dutch Act on the Collective Settlement of
Mass Damage Claims (Wet Collectieve Afwikkeling Massaschade, WCAM) was created in
order to facilitate consensual settlements and grant them binding effect on the relevant victims
affected by the infringement in an opt-out manner. The WCAM entered into force through
amendments to the Civil Code (Art. 7:907-7:910) and Code of Civil Procedure (Art. 1013138
It covers injunctions, declaratory judgments, performance of a contractual duty, termination or rescission of
contract.
139
F. Weber, The Law and Economics of Enforcing European Consumer Law: a Comparative Analysis of
Package Travel and Misleading Advertising (Ashgate Publishing 2014), p. 169.
140
Dutch Minister of Security and Justice, ‘Dutch Draft Bill on Redress of Mass Damages in a Collective
Action’, 7 July 2014.
52
1018) in July 2005, and has subsequently been amended. Since its introduction, the WCAM
has been used to reach a number of successful settlements,141 however, it has not yet been
used for a competition damages claim. It has proven to be an attractive device for
international settlements due to its unique features and bold jurisdictional application that
facilitates the broadest possible victim coverage. According to interviews with stakeholders
from some of the aforementioned settlements, the initiative to start the WCAM procedure
originated from the infringing undertakings.142 To balance its incentivizing application, the
WCAM includes several fairness safeguards encompassed in the judicial review, notification
requirements and eligibility conditions for representative entities.
In order to negotiate a settlement on behalf of the victims, one or more representative entities
need to be established. The representatives can be foundations or associations that are
connected to the collective interests of the victims.143 Moreover, the Consumer Authority can
represent victims that are consumers, 144 but it has not made use of its powers yet.
145
Following the 2013 merger between the Consumer and Competition Authorities, it remains to
be seen whether the merger will impact the private enforcement of competition law in case
the victims include consumers. As for the private representatives, a Dutch ‘foundation’ is
relatively easy and inexpensive to create, and can serve the ad hoc purpose of representing the
victims in the WCAM procedure. An ‘association’ is a more controlled entity that has
members and additional objectives. Both representative entities must be non-profit, but
receiving third party funding is allowed and the WCAM is silent on the legal counsel’s fees,
which generates some financial incentives. Interestingly, in the Converium settlement, the
court took ‘customary US fee practices’ as a reference point when approving a 20% fee for
the US-based legal counsel, which was considered to be excessive and contrary to Dutch
141
The WCAM has produced the 2006 DES settlement on prenatal pharmaceutical drugs, the 2007 Dexia
settlement on the failure to warn about the risks of securities lease products, the 2009 Vie d’Or settlement on the
bankruptcy of an insurance company, the 2009 Shell settlement on misleading statements concerning their oil
and gas reserves, the 2009 Vedior settlement on insider trading, the 2011 Converium settlement on false
statements about the company’s financial position, the 2014 DES II second settlement and the 2014 DSB Bank
settlement on the violation of a duty of care by a bankrupt bank. In 2016, negotiations were initiated in order to
reach a Volkswagen settlement for losses related to their emissions testing fraud, covering securities publicly
traded outside of the US; see Volkswagen Investor Settlement Foundation 2016.
142
I. Tzankova & D. Hensler, ‘Collective Settlements in the Netherlands: Some Empirical Observations’ in C.
Hodges & A. Stadler (eds), Resolving Mass Disputes: ADR and Settlement of Mass Claims (Edward Elgar
Publishing 2013), p. 98.
143
Dutch Civil Code, Art. 7:907.
144
Not stipulated in the WCAM, but added by Art. 2.6 of the Act on the Enforcement of Consumer Protection
Law.
145
Weber 2014 (n 139), p. 168.
53
public policy by the objecting parties.146 Ultimately, the limited restrictions on representative
entities and the lack of funding regulation in the WCAM can be powerful incentives for
bringing large-scale low-value claims.
Once a consensual out-of-court settlement has been reached, the parties must jointly request
approval from the Amsterdam Court of Appeal, which has exclusive jurisdiction to review the
settlement. One of the reasons for appointing this particular court to determine all WCAM
settlements was the possibility to benefit from the financial expertise of its Enterprise
Division.147 In case the Court disapproves of the settlement, the parties could jointly appeal to
the Dutch Supreme Court, which would act in cassation.148 The settlement should contain a
overview of the damage-causing event, any distinctions between the different categories of
victims, an approximate number of victims, the compensation awarded for each victim per
category and the conditions the victims need to fulfill to receive the compensation.149 When
deciding whether to approve the settlement, the Court considers various elements, for
instance: (1) the ‘reasonableness’ of the compensation amount, (2) the appropriateness of the
representative entity to represent the collective interests of the victims, (3) the number of
victims intended to be covered, (4) whether sufficient security is present for the payment of
the compensation amount, (5) the reasonableness of the conditions under which the victims
can claim their individual compensation amount, (6) whether the interests of the victims
cannot be safeguarded in other ways.150 In case the legal counsel’s fee is extracted from the
settlement, the court would assess it under the general ‘reasonableness’ review.151 The Court
also has additional powers: it can seek the counsel of experts, 152 suggest changes to the
settlement (although the parties are not obliged to accept them)153 and assign the costs of
undertaking the negotiations to one of the parties after the approval of the settlement.154
Key safeguards were established by notifying all ‘interested parties’ (i.e. the non-participating
victims) and providing an opportunity to opt-out at any stage of the proceedings. The
notification is a two-stage process, which requires a notice once the settlement has been
146
Tzankova & Hensler 2013 (n 142), p. 101.
B. Krans, ‘The Dutch Act on Collective Settlement of Mass Damages’ (2014) Global Business &
Development Law Journal Vol. 27(2), 281, p. 295.
148
Krans 2014 (n 147), p. 288.
149
Dutch Civil Code, Art. 7:907(2).
150
Dutch Civil Code, Art. 7:907(3).
151
Tzankova & Hensler 2013 (n 142), p. 100.
152
Dutch Code of Civil Procedure, Art. 1016.
153
Dutch Civil Code, Art. 7:907(4).
154
Weber 2014 (n 139), p. 171.
147
54
submitted for approval and again once the final settlement has been declared binding by the
court. The notification requirements can be modified by the Court, but generally, they must
involve direct contact with known victims (i.e. by mail) and a public notification in the
newspapers for the unknown victims.155 In case the victims reside abroad, the Court would
additionally scrutinize the notice according to international and European instruments.156 As a
minimum standard, the victims have at least 3 months to opt-out after the final notification
has been issued.157 The settlement could be made conditional on an opt-out percentage, so that
it could be terminated in case the coverage turns out to be too small.158 Overall, this two-stage
notification process, coupled with court scrutiny, appears adequate to safeguard the interests
of non-participating victims.
Finally, a decisive element of the WCAM’s attraction is its broad jurisdictional application. In
particular, the 2010 Converium settlement extended its scope significantly, since none of the
liable parties and only 3% of the victims were from the Netherlands.159 As highlighted in
chapter 2 of this thesis, the existing EU private international regime is not adequate for
addressing collective proceedings. However, the Amsterdam Court of Appeal has developed
two interpretations of the existing regime in order to facilitate the cross-border opt-out
application of the settlement. One interpretation considers the infringers to be the 'claimants'
and the victims to be the 'defendants' for the purposes of Brussels I Regulation Recast. Since
the settlement represents a consensual agreement, not adversarial litigation, then the nature of
the claimants and defendants would not depend on the infringement itself, but rather, against
whom the settlement would be enforced. Accordingly, Art. 8(1) of the Brussels I Regulation
Recast could be used to claim jurisdiction over multiple defendants that are not domiciled in
the Netherlands, if the claims are found to be 'so closely connected' that there would be a risk
of irreconcilable separate judgments.160 Another jurisdictional interpretation focuses on the
Netherlands as the 'place of performance' in Art. 7(1) of the Brussels I Regulation Recast. It
would deem all relevant parties to be bound by the 'matters relating to the contract', including
155
Dutch Code of Civil Procedure, Art. 1017(3).
Regulation (EC) No 1393/2007 of the European Parliament and of the Council of 13 November 2007 on the
service in the Member States of judicial and extrajudicial documents in civil or commercial matters, OJ L 324;
Convention on the Service Abroad of Judicial and Extrajudicial Documents in Civil or Commercial Matters
(Hague, 1965).
157
Dutch Civil Code, Art. 7:908(2).
158
W.H. van Boom, ‘Collective Settlements of Mass Claims in the Netherlands’ (2009) Rotterdam Institute of
Private Law Accepted Paper Series, p. 15.
159
H. Van Lith, The Dutch Collective Settlements Act and Private International Law (Maklu 2011), p. 22.
160
Tang 2015 (n 121), p. 321-322.
156
55
the compensation stipulated in the settlement.161 For victims that are domiciled outside of the
EU, Art. 3 of the Dutch Code of Civil Procedure is used to declare a 'sufficiently close
connection' to the Dutch legal order.
4.1.2 Public enforcer capacities
In order to gain a better understanding of the Dutch regulatory enforcement architecture
beyond the private enforcement toolbox, the following section will outline additional
elements that can indicate the public enforcer’s capacities to facilitate collective ADR. The
relevant public enforcer is the Netherlands Authority for Consumers and Markets (ACM) that
was created following the merger between the Consumer Authority, Competition Authority
and Independent Post and Telecommunications Authority on 1 April 2013.
Treatment of voluntary compensation - According to the ACM Fining Policy Rule, voluntary
compensation is regarded as a mitigating circumstance that may be taken into account when
setting a fine. Art. 2.10 stipulates that the infringer 'providing full compensation to the parties
injured' is considered to be such a circumstance.162 The wording of the article suggests that
only ‘full compensation’ is covered, however, no further guidance is provided. In practice,
several instances of voluntary compensation can be identified: The 2005 Interpay case
featured two voluntary redress schemes that resulted in an 18% fine reduction. Interpay is a
joint venture for a PIN-transaction network service between Dutch banks and functions as the
sole provider, which compels retailers to buy their services to use PIN payments with their
customers. 163 In 2004, the Competition Authority established that Interpay abused its
dominance by charging excessive prices and that, as shareholders, the Dutch banks were
engaging in an agreement and/or a concerted practice with the object of preventing, restricting
or distorting competition. 164 After the 2005 administrative appeal, the parties negotiated a
structured resolution, which included: (1) the creation of a €10 million fund by donating to
the Foundation for the Promotion of Efficient Payments (Stichting Bevorderen Efficiënt
Betalen), and (2) the creation of a redress scheme that offers the retailers a 1 cent discount for
161
Van Lith 2011 (n 159), p. 50.
Dutch Minister of Economic Affairs, ‘Policy rule of the Minister of Economic Affairs, no. WJZ/14112617, on
the imposition of administrative fines by the Netherlands Authority for Consumers and Markets’, 4 July 2014.
163
M. Bredenoord-Spoek, ‘The Dutch Competition Authority Considers Cooperation and Joint Selling of
Network Services for PIN-transactions to be an Infringement of the Cartel Prohibition (Interpay)’ (2005) eCompetitions Bulletin Excessive Prices Art. No. 35979.
164
Superunie vs Interpay, Dutch Competition Authority, 28 April 2004, Case no. 2978/56.
162
56
every PIN transaction. In return, they received a fine reduction (from €17 million to €14
million) regarding the cartel and the abuse of dominance charges were dropped. 165 In
addition, the 2004-2005 Construction Cartel case featured a voluntary redress scheme that
resulted in a fine reduction amounting to 10% of the compensation amount paid, with an
upper limit of a 10% reduction of the total fine amount. 166 More than 1300 construction
undertakings were involved in a public tender bid-rigging cartel. Approximately 90% of the
undertakings agreed to a fast-lane procedure (resulting in a 15% fine reduction) and
approximately 25% of the undertakings also benefitted from leniency. Interestingly, the victim
that received the compensation was the Dutch Government (including central and local
governments), which could have played a role in the development and approval of this redress
scheme.
Availability of negotiation procedures - Voluntary compensation has often occurred in the
context of negotiation procedures between the public enforcer and the infringers. The
procedural flexibility to ‘negotiate’ is largely derived from Dutch administrative law.167 In
addition, a formal procedure for commitment decisions was introduced through legislative
amendments in Chapter 5A of the Dutch Competition Act in 2007. If successful, negotiations
between the ACM and the undertakings would result in an enforceable formal decision and
the termination of the investigation. 168 Cartel settlement procedures, which result in an
infringement decision and a fine reduction, are formally not available.169 However, similar
settlements were facilitated in the context of a ‘fast-lane procedure’, which required the
undertakings to waive certain procedural rights and agree not to contest the facts, legal
assessment and fines. The procedure resulted in fine reductions, which were given in addition
to any reductions received through leniency. However, this procedure has not been codified in
the relevant instruments nor has it been used in the same manner since its initial development
for the construction cartel.
165
Authority for Consumers and Markets, ‘Press Release: NMa Reviews Fines Imposed on Banks and Interpay’,
22 December 2005.
166
O.W. Brouwer, ‘Antitrust Settlements in the Netherlands: A Useful Source of Inspiration?’ in C.D.
Ehlermann & M. Marquis (eds), European Competition Law Annual 2008: Antitrust Settlements under EC
Competition Law (Hart Publishing 2010), p. 493.
167
P. Kalbfleisch, ‘The Dutch Experience with Plea Bargaining/Direct Settlement’ in C.D. Ehlermann & M.
Marquis (eds), European Competition Law Annual 2008: Antitrust Settlements under EC Competition Law (Hart
Publishing 2010), p. 481-482.
168
Kalbfleisch 2010 (n 167), p. 485.
169
E. Lachnit, ‘Alternative Enforcement of Competition Law – Balancing Legal Requirements in Practice’
(2013) RENFORCE Working Paper, Utrecht Centre for Regulation and Enforcement in Europe, para. 3.2.
57
Treatment of self-regulatory compliance -
The promotion of self-regulatory compliance
efforts can be indicative of the openness to facilitate compliance-oriented enforcement
strategies. The ACM Fining Policy Rule does not list self-regulatory compliance programmes
or other compliance efforts as relevant mitigating circumstances. It has also not published any
guidelines on what well-functioning compliance programme might look like. However, in
general terms, the ACM has promoted the development of self-regulatory compliance systems
by undertakings, since it deems them to be a suitable mechanism for avoiding competition
infringements altogether. 170 Compliance programmes have also been accepted in several
commitment decisions.171 Interestingly, compliance programmes have received more explicit
attention in the telecommunications sector, which now falls under the supervision of the ACM
since the 2013 merger. 172 For instance, the 2006 KPN case regarding illegal customer
discounts featured fine reductions for both voluntary compensation and the creation of a
compliance programme. Moreover, the regulator essentially co-developed the compliance
programme together with KPN. These actions reflected the regulator’s desire to focus more
on the ex ante prevention of breaches and to build ‘high trust’ with the regulatees. 173
Ironically, KPN was subsequently found to have infringed again and this time around the
regulator used the existence of the compliance programme as an aggravating measure to
increase the fine.174
4.2 The United Kingdom
4.2.1 Toolbox of collective proceedings
As of 1 October 2015, the UK features the first tailor-made EU competition enforcement
architecture that integrates collective ADR. This section will provide an overview of the three
devices, which were introduced by the Consumer Rights Act 2015 as amendments to the
170
A. Ottow, Market and Competition Authorities: Good Agency Principles (Oxford University Press 2015), p.
178; statements confirming the ACM’s compliance position stem from C. Fonteijn’s speech on compliance
before the International Chamber of Commerce on 24 January 2014.
171
E. Lachnit, ‘Compliance Programmes in Competition Law: Improving the Approach of Competition
Authorities’ (2014) Utrecht Law Review Vol. 10(5), 31, p. 36.
172
Previously, telecommunications were governed by the Independent Post and Telecommunications Authority
(OPTA).
173
Independent Post and Telecommunications Authority, ‘OPTA Annual Report and Market Monitor 2007’, The
Hague, 27 May 2008, p. 13-14.
174
Lachnit 2014 (n 171), p. 36.
58
Competition Act 1998:175 (1) opt-in/opt-out collective litigation, (2) opt-in/opt-out collective
settlements, and (3) voluntary redress schemes. It is important to note that these devices are
sector-specific to competition law infringements. In addition to the devices above, there are
general collective joinder procedures (Group Litigation Order, representative joinders, test
claim joinders) that bundle individually brought claims.176 However, in case of large-scale
low-value claims, the existence of the joinder device would not create significant incentives
for the voluntary initiation of collective ADR, since the victims are unlikely to bring
individual claims in the first place.
Collective litigation
The collective litigation device housed in Section 47B of the Competition Act did not feature
an opt-out model prior to the 2015 amendments. However, the opt-out model was introduced
with two restrictions: (1) the Competition Appeal Tribunal has discretion to decide whether
the collective claim should proceed on an opt-in or opt-out basis, and (2) the opt-out only
applies for victims domiciled in the UK, whereas non-domiciled victims must explicitly optin. 177 The claimants could be consumers and businesses, although the policy discussions
preceding the Consumer Rights Act were largely geared at consumers. Ultimately, including
businesses in collective litigation was found to be especially helpful for SME-s.178
In order to litigate, a representative entity would have to be established. The Court would only
authorise representative entities if they consider their appointment to be ‘just and reasonable’,
taking into account: their capacity to act ‘fairly and adequately’ in the best interests of the
class, whether it is a pre-existing entity, whether it is a member of the class itself, its ability to
meet the costs of the proceedings and the existence of conflicts of interest.179 The CAT Rules
do not contain any further limitations, which suggests that law firms and special purpose
vehicles can be representative entities as well, although their exclusion was explicitly
175
The Consumer Rights Act entered into force on 1 October 2015. The new regime applies to claims that were
raised from that date onwards, even if the infringement took place earlier. While the amended provisions are
brief, detailed information can be found in two additional sources: the Competition Appeal Tribunal Rules 2015
and the Competition and Markets Authority’s Guidance on the approval of voluntary redress schemes for
infringements of competition law.
176
Civil Procedure Rules 19.11, 19.6, 19.15.
177
Competition Act 1998 47B(4)-(7), 47B(11); Competition Appeal Tribunal Rules 2015 79(3).
178
UK Department for Business Innovation and Skills, ‘Private Actions in Competition Law: A Consultation on
Options for Reform - Government Response’, January 2013, section 5.19.
179
CAT Rules 78(2), 78(3).
59
suggested by the Government in its response during the policy discussions.180 If discretion
was indeed left for the Court in this regard, then the creation of financial incentives for
representing large-scale low-value claims could be ensured.
The Court would then determine whether the presented claims raise ‘the same, similar or
related issues of fact or law’, which were also defined in the CAT Rules as ‘common
issues’.181 The suitability of joining the claims would depend on the predicted cost and benefit
ratio, the size and nature of the class, the availability of ADR, and the existence of CMAapproved voluntary redress schemes or any other attempts at voluntary compensation. 182
Approving an opt-out application would additionally depend on an assessment of the strength
of the claims and the individual damages amounts that victims could retrieve through an optin application. 183 Overall, the standards of scrutiny appear to be very responsive to the
circumstances of the case (i.e. sensitive to the difficulties of large-scale low-value claims) and
create direct incentives for the initiation of collective ADR by taking the voluntary
compensation efforts into account.
If the collective proceedings were successful, then the compensation would be made available
for the victims. In case an opt-out model applied, the Court would have the obligation to
award damages to the representative entity or any other person that it ‘thinks fit’, whereas in
case of opt-in, it would have the discretion not to do so.184 Any unclaimed damages resulting
from an opt-out model could have two outlets: a cy-près donation to the Access to Justice
Foundation or the assignment of a part or all of the unclaimed damages to the representative
entity to cover litigation costs.185 Considering that contingency fees and exemplary damages
are prohibited,186 retaining the possibility of obtaining parts of the damages would reduce the
level of risk involved. Ultimately, the provisions appear to strike a careful balance between
safeguards and financial incentives, but due to the level of discretion left to the Court, it
remains to be seen how the device will function in practice.
180
BIS 2013 (n 178), section 5.32.
CA1998 47B(6); CAT Rules 73(2). It appears to be an easier standard to satisfy than the ‘same interest’
standard under the Civil Procedure Rules. See Rodger 2015 (n 7), p. 274; Civil Procedure Rules 19.6(1).
182
CAT Rules 79(2).
183
CAT Rules 79(3).
184
CA1998 47C(3).
185
CA1998 47C(5), 47C(6).
186
Exemplary damages prohibited: CA1998 47C(1). Contingency fees appear to be prohibited only in case of an
opt-out application: CA1998 47C(8); Courts and Legal Services Act 1990 section 58AA(3).
181
60
Collective settlements
The collective settlement regime features two distinct devices: (1) The first device is a section
49A settlement, which only applies in the context of ongoing collective litigation. Once the
collective litigation proceedings have been initiated, then resorting to ‘regular’ settlements
outside of the framework of this regime would be prohibited.187 The Government justified this
exclusion as a way of preventing the defendant from making low settlement offers. Concerns
would arise if the claimants were to reject a low settlement offer and then become liable for
litigation costs in case they lost. 188 The Government found the more regulated settlement
framework to be better suited for protecting the interests of the victims. 189 The regulated
framework represents a tradeoff that protects the interests of representative entities and
ultimately still creates incentives for bringing collective claims. (2) The second device is a
section 49B settlement, which can apply even if no collective litigation has been started. At
first sight, the utility of introducing a second judicially-approved settlement device could
appear questionable.190 However, this settlement device is mainly meant to facilitate voluntary
compensation efforts that have arisen without the threatening existence of a certified
collective litigation claim. It would be counterintuitive to only make opt-out settlements
available in the context of litigation - especially in cases, where all of the parties have reached
a consensual outcome, but the infringers would have to be formally sued before the parties
have access to a settlement approval process.191
An application for settlement approval could either be made after a litigation order has been
issued, in case of a 49A settlement, or at an earlier stage, in case of a 49B settlement.192 In the
first scenario, by issuing the litigation order, the CAT has already determined that the class
has an eligible claim that meets the conditions regarding the strength of the claim, the
suitability of the representative entity, the size and nature of the class etc. In the second
scenario, lacking an existing order, the CAT would undertake a similar certification process
187
CAT Rules 94(1).
UK Department for Business Innovation & Skills, ‘Competition Appeal Tribunal Rules of Procedure: Review
by the Right Honourable Sir John Mummery: Consultation’, February 2015, section 7.9-7.12.
189
UK Department for Business Innovation & Skills, ‘Competition Appeal Tribunal Rules of Procedure:
Government Response’, September 2015, section 2.95.
190
Mulheron 2016 (n 136), p. 19.
191
Mulheron 2016 (n 136), p. 20; the Government stated in its response that ‘it would be perverse to require a
company to effectively invite someone to sue them in order to reach a settlement’, BIS 2013 (n 178), p. 6.19.
192
CA1998 49A(1), 49B(1).
188
61
oriented towards a settlement.193 The application for approval must be made together by the
defendant and the representative entity. 194 It must contain a statement stipulating that the
settlement terms are ‘just and reasonable’ with supporting evidence by experts, as well as
details of the claims covered, notification arrangements for non-participating victims and
distribution plans for the compensation.195 The Court could follow up with directions, after
which a hearing will be held and the settlement terms will be scrutinized. The settlement
would have to meet the ‘just and reasonable’ review standard, which covers the damages
amount offered, the number of persons covered, the expert opinions, the duration, cost and
chances of success with collective litigation, especially if it would be possible to obtain
compensation amounts ‘significantly in excess’ of what the settlement offers.196
If the settlement is approved, then the compensation will be made available to the victims
according to the terms of the settlement. There appear to be no provisions prescribing the
treatment of unclaimed settlement funds.197 However, the plans for unclaimed funds are taken
into account in the ‘just and reasonable’ scrutiny. Interestingly, the possibility of reverting
funds back to the defendant is explicitly allowed.198
Voluntary redress schemes
The voluntary redress schemes, housed in Sections 49C, 49D and 49E of the Competition Act,
stand separate from the two devices mentioned above, since their existence is not contingent
on the victims initiating collective litigation. In essence, this form of voluntary compensation
can be viewed as an infringer-led out-of-court settlement, which requires the victims to opt-in
by accepting the settlement amount. In other words, the settlement does not have res judicata
effect on all relevant victims in an opt-out manner. The schemes are also not overseen by the
CAT, but instead by the CMA or other sector-specific regulators with competition powers.199
A scheme can be approved simultaneously with the issuing of the CMA infringement decision
193
CA1998 49B(5); CAT Rules 96(6), (9)-(11).
CA1998 49A(2), 49B(2).
195
CAT Rules 94(4), 97(2).
196
CAT Rules 94(9), 97(7).
197
Rodger 2015 (n 7), p 284.
198
CAT Rules 94(9)(g), 97(7)(g).
199
Competition and Markets Authority, ‘Guidance on the Approval of Voluntary Redress Schemes for
Infringements of Competition Law’, London, 14 August 2015, 1.1, fn. 3 p. 2. For the purposes of the discussions
that follow, the CMA will be used as the main example.
194
62
or as a follow-on initiative in case of pre-existing infringement decisions by the CMA or
European Commission.200
The most significant incentive for initiating a voluntary redress scheme is the possibility of
receiving up to 20% fine reductions. 201 This option is only available in the context of
upcoming CMA infringement decisions and subject to the CMA’s discretion, although it states
that a fine reduction would be awarded ‘in the majority of the cases’.202 If the applicant has
proposed a scheme that the CMA would strongly consider approving, then the early intention
to award a fine reduction would be mentioned in the draft penalty statement, making it visible
for all parties to the investigation. When deciding the fine reduction percentage, the CMA
would consider the terms and administrative costs of the scheme, the size of the penalty, as
well as the existence of leniency and settlements. However, it would not consider issues such
as the nature and gravity of the infringement.203
Remarkably, the task of devising the scheme was granted to third parties: the Chairperson and
the Board. The applicant can only appoint the Chairperson, who should be a senior lawyer or
judge with sufficient demonstrable knowledge of competition law. The Chairperson would
then appoint the members of the Board, who should include an economist, an industry expert,
a person representing the interests of the beneficiaries of the scheme (e.g. a consumer body,
trade association) and any other person suitable due to their expert knowledge. 204 The
applicant would pay for their services and the payment could not be contingent on the
outcome of the final assessment of the scheme.205 The Chairperson and the Board would have
to meet standards of independence, impartiality, objectivity, integrity and honesty; and
perform their duties with reasonable skill and care, in accordance with the law and rules
relating to their professional conduct. This means that any actual or potential conflicts of
interest, which might cause them to favour the applicant or the beneficiaries, are prohibited
and would have to be disclosed.206 The Chairperson and the Board are expected to engage in
extensive fact-finding and cooperation with the applicant, while respecting confidentiality
200
CA1998 49C(2).
CMA Guidance (n 199), 3.30.
202
CMA Guidance (n 199), 3.29.
203
CMA Guidance (n 199), 3.31.
204
CMA Guidance (n 199), 2.43-2.45.
205
CMA Guidance (n 199), 2.47.
206
CMA Guidance (n 199), 2.49-2.59.
201
63
with regards to sensitive information.207 They are expected to work within the parameters and
wishes suggested by the applicant, but they are allowed to go beyond them.208 Ultimately, it is
up to the Chairperson and the Board, not the applicant, to devise a comprehensive redress
scheme that addresses the nature of the potential beneficiaries (e.g. whether indirect
purchasers would be covered), the types of loss that are recoverable, the type of compensation
offered (whether it is monetary or non-monetary compensation; although a choice between
the two is insisted on by the CMA), the process of applying for compensation, the evidence
requirements, the existence of a complaints process in case of the rejection of a beneficiary’s
claim, the advertisement plans (in addition to the publication of the schemes on the
Government’s websites), and the legal consequences of accepting compensation under the
scheme.209
Although the CMA has the power to grant final approvals, it is not meant to carry out a
detailed external review of the scheme, especially when it would ‘duplicate or undermine’ the
work of the Chairperson and the Board. 210 In its review, the CMA can consider the
compensation amount offered, the duration of the scheme, advertising, evidence requirements,
the legal consequences of the acceptance of redress, the position of vulnerable consumers, the
objectivity and independence of the Chairperson and the Board, as well as any dissenting
views from the applicant, the Chairperson or the Board. 211 Although the initiation of the
scheme is voluntary, once the CMA gives its approval, the compensating party is under a legal
duty to comply. The duty is owed to the beneficiaries of the scheme, who can bring civil
proceedings in case of non-compliance.212 The schemes that relate to upcoming infringement
decisions can be approved conditionally on the basis of an outline, subject to providing
further details about the redress scheme in the future. Even if conditional approval was
granted, the CMA reserves the right to approve the same scheme under other conditions,
approve a different scheme as a replacement or withdraw the approval altogether. 213
Interestingly, the CMA was also granted the power to recover ‘reasonable costs’ for its
services in facilitating the voluntary redress schemes. It will be interesting to see how the
207
CMA Guidance (n 199), 2.75-2.78.
CMA Guidance (n 199), 2.71.
209
CMA Guidance (n 199), 2.11-2.41.
210
CMA Guidance (n 199), 3.5.
211
CA1998 49C(3).
212
CA1998 49E(1)-(4).
213
CA1998 49C(4) & 49C(5)(b).
208
64
CMA approaches its cost recovery powers in practice, especially since it is allowed to revoke
the entire scheme in case of non-payment.
4.2.2 Public enforcer capacities
In order to gain a better understanding of the UK regulatory enforcement architecture beyond
the private enforcement toolbox, the following section will outline additional elements that
can indicate the public enforcer’s capacities to facilitate collective ADR. The relevant public
enforcer is the Competition and Markets Authority (CMA) that was created following the
merger between the Office of Fair Trading and Competition Commission on 1 April 2014.
Treatment of voluntary compensation - As already outlined in the previous section, the new
voluntary redress scheme framework constitutes the first highly regulated attempt at inducing
voluntary compensation. Outside of this framework, the fining guidelines do not list voluntary
compensation as a mitigating factor. However, in practice, compensation has been
acknowledged as a relevant circumstance on rare occasions. For instance, the 2006
Independent Schools case featured a voluntary redress scheme that resulted in a fine
reduction. 214 The case involved fifty non-profit schools that engaged in an information
exchange regarding the fixing of tuition fees for upcoming years. Their regular and systematic
exchange constituted as an agreement or a concerted practice with the object of preventing,
restricting or distorting competition. After the Statement of Objections was published, the
schools engaged in negotiations with the OFT and reached the following resolution:
admittance to the infringement, an ex gratia payment of £3 million to an educational trust
fund for the benefit of the students that attended the schools during the infringement, and a
reduced fine of £10 000 for each school.215 Next to inducing voluntary compensation, it was
also the first time that the public enforcer used a cartel settlement procedure. The OFT Chief
Executive noted that the decision illustrated their responsiveness to 'consider innovative
solutions' in special cases. 216 With the 2015 reforms that introduced regulated ‘voluntary
redress schemes’, it remains to be seen how much room was left to develop any additional
and innovative compensatory outcomes.
214
UK Independent Schools (2006) OFT Decision No. CA98/05/2006.
UK Independent Schools (n 214), para. 36. Some schools received further up to 50% reductions for leniency.
216
Office of Fair Trading, ‘OFT Issues Final Decision and Imposes Penalties in Independent Schools
Investigation’, 24 November 2006.
215
65
Availability of negotiation procedures - Considerations for voluntary compensation could
emerge in the context of negotiation procedures between the public enforcer and the
infringers. Commitment decisions were introduced through legislative amendments in Section
31A of the Competition Act in 2004. The accompanying guidance outlines the circumstances
under which the enforcer may consider accepting commitments. If successful, the procedure
would result in an enforceable formal decision and the termination of the investigation.217 In
addition, following the Independent Schools case discussed above, various settlement-type
procedures (often referred to as ‘early resolution’) were informally applied. More detailed
guidance was introduced in 2014, outlining the specifics of the formal settlement procedure.
The settlement requires, among other things, a voluntary admission of guilt and could result
in up to a 20% or 10% fine reduction, depending on whether the procedure occurred before or
after the Statement of Objections was issued.218 These reductions would be in addition to any
reductions received through leniency.219
Treatment of self-regulatory compliance - Self-regulation has enjoyed a prominent place
within the enforcement strategies of various UK sector-specific regulators and has been
systematically promoted by the government.220 To some extent, the CMA also reflects this
policy approach by acknowledging and actively promoting the development of self-regulatory
compliance programmes by the undertakings. The existence of well-functioning compliance
programmes or other 'adequate steps' to ensure compliance are deemed to be mitigating
factors that can result in up to a 10% fine reduction.221 However, the mere existence of a
compliance programme would not suffice. The undertaking would have to prove that they
took the necessary steps to ensure the functioning and maintenance of the programme.
Interestingly, the CMA reserves the right to use the existence of a compliance programme as
an aggravating factor in exceptional cases, where, for example, the undertaking used it to
deceive the enforcer and conceal an infringement. 222 The CMA has published extensive
guidance documents on how to ensure compliance with competition law and even developed
217
Office of Fair Trading, ‘Enforcement: Incorporating the Office of Fair Trading’s Guidance as to the
Circumstances in Which it May be Appropriate to Accept Commitments’ (OFT407), December 2004, para. 4.
218
Competition Act 1998: Guidance on the CMA's Investigation Procedures in Competition Act 1998 Cases
(CMA8), Competition and Markets Authority, March 2014, para. 14; OFT's Guidance as to the Appropriate
Amount of a Penalty (OFT423), Office of Fair Trading, September 2012, para. 2.26.
219
CMA 2014 (n 218), para. 14.3.
220
Hodges 2015 (n 41), p. 173, 178-182.
221
OFT 2012 (n 218), para. 2.15.
222
OFT 2012 (n 218), p. 12-13.
66
special guides for small businesses. 223 It promotes a ‘four-step risk-based approach’ that
includes: (1) risk identification, (2) risk assessment, (3) risk mitigation, and (4) regular
programme review.224 Its understanding of self-regulatory compliance programmes is in line
with the behavioural findings and organisational studies that view a top-down commitment to
compliance as an essential component for the development of a compliance culture. 225 In
2015, the CMA published a study on the undertakings’ understanding of competition law,
including their perception of the most important reasons for compliance and their awareness
of the consequences of non-compliance.226 According to the findings, undertaking of all sizes
are not as familiar with competition compliance issues as expected, which would suggest that
the current guidance might not be effective.227
4.3 Comparative analysis
Having obtained a sufficiently detailed overview of both national architectures, this section
will continue with a comparative analysis. In particular, a selection of key issues will be
analyzed in more depth: (1) the incentives that the architecture provides to initiate collective
ADR, (2) the funding incentives for representative entities, (3) the structural differences
between the two architectures, (4) the fairness review that the compensatory outcome is
subjected to, and (5) the role of the public enforcer in facilitating collective ADR. Before
starting the analysis, some preliminary consideration will be given to certain common
denominators between the Netherlands and the UK, which could affect the broader
discussions on the introduction of a collective ADR model: their general attitude towards
ADR in civil justice and their possible aspiration to attract international legal activity.
Firstly, the presence of ADR in civil justice systems is notable in both the Netherlands and the
UK. The Dutch dispute resolution landscape is illustrated by the large percentage of out-ofcourt resolutions and the proliferation of ADR entities, which reflects the alleged cultural
223
Office of Fair Trading, ‘Guidance: How Your Business Can Achieve Compliance with Competition Law’
(OFT1341), June 2011; CMA 2014 (n 218).
224
OFT 2011 (n 223), p. 9.
225
Ottow 2015 (n 170), p. 184.
226
IFF Research, ‘UK businesses’ understanding of competition law, Prepared for the CMA’, 26 March 2015.
227
B. Rodger, ‘Competition Law Compliance: the CMA 2015 Study, Compliance Rationales and the Need for
Increased Compliance Professionalism and Education’ (2015) European Competition Law Review Vol. 36(10),
423, p. 430.
67
tendency towards conflict-avoidance and negotiated settlements. 228 The Dutch government
has promoted ADR in order to prevent unnecessary burdens on civil courts and develop a
more efficient justice system. 229 The courts encourage settlements and regularly make
voluntary referrals to mediation. Moreover, there is a national network of Consumer
Complaint Boards (Geschillencommissie) that provide dispute resolution services in B2C
situations, which result in a binding outcome for the undertaking. However, these ADR
entities have not yet managed collective cases. Finally, arbitration is generally used for
industry and trade disputes, partly as a result of the existence of renowned sector-specific
institutions.230 In the UK, Lord Woolf’s Access to Justice review and the subsequent 1999
reforms of the civil justice systems in England and Wales ushered in a significant emphasis on
ADR. The courts were tasked with various case-management powers, coupled with the
objective of promoting consensual dispute resolution throughout the civil proceedings231 and
the use of ADR was directly incentivized by taking its use into account when deciding the
amount of costs related to the litigation process.232 Moreover, many disputes are handled by
the numerous sector-specific ADR entities that can impose a binding outcome on the
undertakings, which is illustrative of the broader encouragement of self-regulation in the
UK. 233 Self-regulation, coupled with ADR mechanisms, is strengthened by statutory
provisions, supervision by public regulators, codes of conduct, government funding and
education.234 Arbitration is generally used for industry and trade disputes, and the UK houses
several important international arbitral entities. Overall, it is likely that the favourable attitude
towards ADR in civil justice could have a certain impact on the functioning and potential
success of collective ADR in both countries.
Secondly, it has been suggested that the Netherlands and the UK aspire to become the
hotspots for international legal activity.235 If this hypothesis is true, then it could explain and
228
F. Weber & C. Hodges, ‘The Netherlands’ in C. Hodges, I. Benöhr & N. Creutzfeldt-Banda (eds), Consumer
ADR in Europe (Hart Publishing 2012), p. 130; for a reference to the Dutch ‘conciliatory’ legal culture, see T.
Arons & W.H. van Boom, 'Beyond Tulips and Cheese: Exporting Mass Securities Claim Settlements from The
Netherlands' (2010) European Business Law Review Vol. 21(6), 857, p. 883.
229
M. Pel, ‘Regulation of Dispute Resolution in the Netherlands: Does Regulation Support or Hinder the Use of
ADR?’ in F. Steffek & H. Unberath (eds), Regulating Dispute Resolution: ADR and Access to Justice at the
Crossroads (Hart Publishing 2013), p. 299.
230
Pel 2013 (n 229), p. 302.
231
Lord Woolf 1996; Civil Procedure Rules.
232
Civil Procedure Rules, Rule 44.4(3)(a)(iii).
233
For an extensive overview, see Hodges, Benöhr & Creutzfeldt-Banda 2012 (n 115), p. 253.
234
Hodges, Benöhr & Creutzfeldt-Banda 2012 (n 115), p. 338.
235
G. Wagner, ‘Mass Tort Resolution: Competition Between Jurisdictions and Mechanisms’ in W. van Boom &
G. Wagner (eds), Mass Torts in Europe: Cases and Reflections (Walter de Gruyter 2014), p. 293-294.
68
influence the development of particular models of collective redress. Even after the 2013
Commission Recommendation on Collective Redress, the European landscape of collective
proceedings is diverse and a predominant model cannot be said to have emerged yet.
Moreover, the fragmentation is fueled by the lack of a tailor-made EU private international
law regime for collective proceedings. Against this background, regulatory competition could
occur between different national regimes, as well as between public and private devices.236
The choice between different national regimes could include various parameters, such as the
efficiency and reputability of the civil justice system as a whole. However, when choosing an
avenue for collective redress, particular parameters could be determinative. Firstly, in order to
attract collective legal activity from other jurisdictions, the availability of collective devices
with a broad scope is essential. Secondly, the facilitation of an opt-out model could provide a
competitive advantage, which enables infringing undertakings and representative entities to
achieve once-and-for-all solutions by providing the broadest coverage of victims. Thirdly,
given the low cost, speed and flexibility of ADR mechanisms, the creation of collective ADR
devices could be very attractive for undertakings that can be incentivized to offer
compensation proactively. Adopting all three of the aforementioned measures is likely to
indicate the willingness and desire to facilitate international legal activity in the form of
collective redress.
Overview
To recap, the core components of both regulatory enforcement architectures are summarized
in the table below with collective ADR devices highlighted in grey:
236
H. Eidenmüller (ed), Regulatory competition in contract law and dispute resolution (Hart Publishing 2013),
p. 5.
69
Architectural component
Collective litigation
UK
Netherlands
Yes - opt-out/opt-in collective
No. However, opt-out
litigation subject to the
collective litigation exists,
discretion of the CAT.
but does not provide
Competition-specific regime.
‘damages’ as a remedy.
Opt-out application limited to
UK domiciled victims.
Collective settlements
Yes - opt-out/opt-in collective
Yes - opt-out collective
settlement.
settlement. Opt-out
Opt-out application limited to
application worldwide.
UK domiciled victims. Section
49A settlement only applicable
in case of ongoing collective
litigation. Section 49B
settlement applicable for other
scenarios.
Regulated voluntary
Yes. Schemes function like an
compensation:
opt-in settlement. Up to 20%
voluntary redress schemes
fine reduction.
Other voluntary
No, formally not a mitigating
Yes, mitigating factor leading
compensation
factor.
to fine reductions.
Self-regulatory compliance
Yes, mitigating factor leading
No, formally not a mitigating
to fine reductions.
factor.
Yes, commitments and
Yes, commitments.
settlements.
Settlements formally not
Negotiation procedures
No.
available.
70
The comparative analysis now shifts to a more in-depth assessment of the five key categories
that are deemed most crucial for the success of the collective ADR model:
(1) Structural incentives for collective ADR
In the Netherlands, collective ADR was introduced through opt-out collective settlements.
The WCAM works well in situations, where the infringer finds the achievement of once-andfor-all opt-out settlements to be attractive due to its goodwill or in case the incentives are
provided by external sources. The latter scenario can arise if the infringer is facing collective
litigation in another jurisdiction or individual litigation in multiple jurisdictions. If the
infringer is cooperative, the WCAM would be an excellent device, since the resulting
settlement would be court approved and thereby secure legal certainty insofar as possible
under the existing private international law regimes. The jurisdictional claims asserted
through the WCAM have been particularly bold, covering international victims and
infringers. However, in case the infringer is not cooperative, then the WCAM might not create
sufficient incentives on its own. As a public trigger, the public enforcer provides direct
incentives through fine reductions, since it considers compensation efforts as mitigating
factors when setting the fine. However, there is a notable lack of a relevant opt-out collective
litigation device. Amending their existing opt-out collective litigation device by adding the
possibility to obtain damages as a remedy, would change the situation and be likely to create
sufficient incentives to initiate collective ADR.
In the UK, collective ADR was introduced through competition-specific opt-out collective
settlements and voluntary redress schemes. Although initiating collective ADR is entirely
voluntary, strong private and public trigger incentives were set in place. The biggest incentive
for early voluntary compensation is the possibility to receive up to 20% fine reductions.
Moreover, collective ADR was carefully integrated into the broader collective proceedings
framework. The introduction of opt-out collective litigation can be a significant incentive, as
long as the victims and their representatives have access to sufficient funding in order to
materialize the threat of litigation. However, the impact of the opt-out model is somewhat
diluted by restricting its automatic application to UK residents, while the rest of the victims
would have to opt-in. Voluntary compensation efforts, either privately or through the CMAapproved redress schemes, have a direct impact on any subsequent collective proceedings,
since the initial application would have to include a statement whether the parties have
71
attempted to use ADR. This would bring their ADR history (or lack thereof) to the attention of
the CAT right from the beginning of the proceedings. Collective ADR is thereafter taken into
account when deciding whether to authorise collective proceedings on an opt-out basis,
whether particular claims are eligible for the collective proceedings, and when determining
the amount of recoverable litigation costs. Overall, as long as the likelihood of collective
litigation is considerable, the UK architecture is capable of creating sufficient private and
public triggers to initiate collective ADR.
(2) Funding of representative entities
The structural incentives outlined above were primarily directed at inducing the infringing
undertaking’s initiation of collective ADR. Once the infringer’s consent is ensured, then the
offer to start collective ADR proceedings or to receive a complete redress scheme can be
incentivizing enough for the representative entities or their funders due to the high certainty of
receiving their share of the compensation. However, particularly in case the threat of
collective litigation needs to be established in order for collective ADR to be induced in the
first place, additional financial incentives for representative entities could be needed.
The UK collective litigation device allows law firms and special purpose vehicles as
representative entities. As a safeguard against the creation of a litigation culture, legal counsel
was not allowed to operate with contingency fees. However, no contingency fee prohibition
was stipulated for third party funders. Moreover, in case of unclaimed compensation
following an opt-out application, the CAT may order that the representative entity could
receive the damages 'in respect of the costs or expenses incurred', which would include the fee
charged by a third party funder. 237 These elements could make the funding of collective
litigation financially incentivizing for third party funders. The Netherlands does not feature a
collective litigation device that provides damages as a remedy. However, the treatment of
funding under the WCAM opt-out settlement device bears a mention in case a similar
approach were to be applied to a ligation device in the future. Notably, the WCAM sets very
few restrictions on representative entities and allows for third party funding. Moreover, the
judicial review of the WCAM settlement does not include the legal counsel's fees as a
separate point of assessment. Only in case the fees are derived from the settlement amount
would they be included in the general 'reasonableness' review of the settlement and, even
237
CA1998 47C(6); Mulheron 2015 (n 71), p. 309
72
then, the court has been very generous in its assessment. It would also not be difficult to
circumvent the assessment of the fees by simply agreeing to an external payment method.238
Overall, the UK and the Netherlands have reasonably facilitated the creation of financial
incentives for bringing collective proceedings by not creating any substantial restrictions.
(3) Structural differences
The scope of collective ADR in the Netherlands and in the UK is different, since the latter
chose to limit the collective redress regime to competition claims only. A competition-specific
collective redress regime was primarily developed due to an empirical need that was
identified in this area, but not in others.239 The UK Government believed that the need for
collective redress, especially using the opt-out model, should be assessed sector-by-sector
following an impact assessment.240 In comparison, the Dutch Government maintains a general
preference for horizontal, cross-sectoral devices. For instance, in its response to the White
Paper on Antitrust Damages, the Dutch Government expressed disappointment with the
choice to adopt a ‘fragmented’ competition-specific approach to collective redress. 241
Moreover, the Dutch Government believed that the Commission failed to fully appreciate the
potential of out-of-court settlements with broad coverage and suggested that the WCAM
could serve as inspiration for future European developments.242 Overall, although the success
of a collective regime would depend on a number of parameters other than its horizontal or
vertical application, the development of some competition-specific rules could facilitate some
flexibility with regards to problems that arise with competition damages claims. For instance,
competition-specific treatment could be needed for the identification of a 'common interest'
between direct and indirect purchasers, and for the approximations in quantifying damages.243
The UK regime does not appear to address these issues explicitly, but the oversight provided
by a specialized competition tribunal could contribute to developing workable outcomes.
238
Tzankova & Hensler 2013 (n 142), p. 101.
The Government’s Response to the Civil Justice Council’s Report: ‘Improving Access to Justice through
Collective Actions’, Ministry of Justice, July 2009, p. 5.
240
Ministry of Justice 2009 (n 239), p. 6.
241
Dutch Government, ‘The Netherlands' Response to the White Paper on Damages Actions for Breach of the
EC Antitrust Rules’, 2008, p. 2. The same point was stressed in response to the Green Paper on Consumer
Collective Redress - Dutch Government, ‘Dutch Response to the Green Paper on Consumer Collective Redress’,
2008, p. 3.
242
Dutch Government 2008 (n 241), p. 3-5.
243
Athanassiou 2014 (n 12), p. 152-155.
239
73
The UK’s opt-out settlement regime is often considered to be ‘inspired’ by the Dutch opt-out
settlement device. Indeed, both settlement regimes share many similarities regarding their
functioning. However, there are in fact two distinct settlement devices in the UK regime: (1)
section 49A settlements, which applies in case of ongoing collective litigation; and (2) section
49B settlements, which can apply in other cases. Section 49A settlements are not comparable
to the Dutch settlements, since they are not available for any type of out-of-court settlement
scenario - they must exclusively take place in the post-certification context of ongoing
collective litigation. Moreover, ‘regular’ out-of-court settlements would be forbidden once
collective litigation is initiated. In comparison, the Dutch opt-out settlement device can be
used in broader scenarios and does not have to be preceded by litigation. In that sense, only
the section 49B settlement device resembles the Dutch settlement device. However, both optout settlement devices under the UK regime are ultimately limited in their opt-out application
- only UK domiciled victims are automatically included, whereas other victims would have to
explicitly opt-in. In comparison, the Dutch device has been applied very generously by the
Amsterdam Court of Appeal, asserting opt-out jurisdiction over EU and non-EU victims alike.
In sum, the Dutch opt-out settlement device has a much broader scope and cannot be
considered an equivalent of the UK settlement devices.
Both the Netherlands and the UK facilitate voluntary compensation efforts. However, the UK
developed a highly regulated framework for ‘voluntary redress schemes’, whereas the
Netherlands did not. The Dutch public enforcer reserved the discretion to consider any types
of voluntary compensation efforts as mitigating factors. On the other hand, creating a
regulated framework for voluntary compensation enables to assert more public control over
the process and thereby ensure fairness guarantees. Moreover, by creating specific guidelines
for the successful development of redress schemes, the infringing undertakings are given
more certainty that their compensation efforts would result in a fine reduction. In order to
ensure that the redress scheme is adequate, the UK framework prescribes the creation of an
independent body of experts (a Chairperson and the Board), which must at least include an
economist, an industry expert and a person representing the interests of the beneficiaries of
the scheme.244 Having the independent experts design the redress schemes can ensure that
both the requirements stipulated by the CMA as well as the expectations of the infringing
undertaking would be reasonably met. Yet, these added layers of complexity could make the
voluntary redress schemes a less economical choice for certain infringing undertakings, who
244
CMA Guidance (n 199), 2.43-2.45.
74
might prefer to engage in direct settlement negotiations. After all, in addition to providing the
compensation amount, the undertaking would also have to pay for the services of the
independent experts. Ultimately, the UK architecture does not exclude the development of
other voluntary compensation efforts, but it only provides fine reductions for the voluntary
redress schemes created in accordance with the regulated framework.
(4) Fairness review
In order to achieve the recognition and enforcement of opt-out settlements, the parties would
have to obtain court approval through judicial review. Both the UK and the Netherlands
subject the settlement terms to scrutiny by the Competition Appeal Tribunal and the
Amsterdam Court of Appeal, respectively.
In the UK, the settlement approval process starts with class certification,245 whereby the CAT
would review whether the representative entity is suitable and whether the claims are eligible
for inclusion in the class.246 In making these determinations, the CAT would consider the
existence of common issues and the broader suitability of aggregation. Importantly, the prior
use of collective ADR would be highlighted when approving the class. Moreover, victim
participation is ensured by allowing them to make a submission either in writing or orally
during the certification hearing. After the settlement has been presented for approval, the CAT
would scrutinize the settlement terms to determine whether they are ‘just and reasonable’ by
taking into account ‘all relevant circumstances’, particularly the following factors: the
compensation amount and settlement terms (including payment of related fees), the number
victims covered, the predicted compensation to be obtained through collective litigation
(especially if the amount would be significantly in excess of the settlement), the predicted
duration and cost of collective litigation, expert opinions, victim opinions, and plans for
unclaimed funds. 247 There appear to be no further provisions prescribing the treatment of
unclaimed settlement funds, aside from allowing for the possibility of reverting the funds
245
Unless the certification has already taken place in the context of ongoing collective litigation.
CAT Rules 96(6), 96(9)-(11), 79.
247
CAT Rules 94(9), 97(7).
246
75
back to the defendant. 248 Overall, the judicial review standards allow for rather intense
scrutiny of all relevant circumstances related to the settlement.
In the Netherlands, the settlement approval process also includes a preliminary stage,
however, it does not resemble a traditional ‘class certification’, whereby some of the
substantive issues surrounding the collective claims would already be assessed. The
representative parties are simply required to make a joint appeal for settlement approval,
which is followed by a court hearing. Victim participation would be ensured through
notification and by allowing them to make submissions during the hearing. When deciding
whether to approve the settlement, the Court would consider various elements, such as: the
‘reasonableness’ of the compensation amount, the appropriateness of the representative entity
to represent the collective interests of the victims (e.g. whether the foundation or association
fills the formal requirement of including the protection of the victims in its articles of
association ), the number of victims intended to be covered, whether sufficient security is
present for the payment of the compensation amount, the reasonableness of the conditions
under which the victims can claim their individual compensation amount, and whether the
interests of the victims cannot be safeguarded in other ways.249 In addition, the Court can
scrutinize the notification arrangements, which have to be effective and ensure that all
international victims are made aware of their right to opt-out.250 As is evident from the review
conditions above, the Dutch review leaves the parties a lot of discretion to mold the
settlement to their wishes.251 In comparison to the UK judicial review standards, the financing
of the settlement is not subjected to detailed scrutiny, although it is partly covered by the
general ‘reasonableness’ review. Moreover, the appropriate treatment of unclaimed settlement
funds is not prescribed, which allows the parties to revert the funds back to the defendant or
design a cy-près solution.252 Finally, in case the Court questions some of the fairness aspects
of the proposed settlement, it can make non-binding suggestions and give the parties a chance
to make amendments before rejecting it.253 Overall, the judicial review scrutinizes the key
248
CA1998 47C(5), 47C(6); CAT Rules 94(9)(g), 97(7)(g). In comparison, unclaimed funds from collective
litigation were specifically highlighting the option of directing funds to the Access to Justice Foundation as a cyprès donation.
249
Dutch Civil Code, Art. 7:907(3).
250
Dutch Code of Civil Procedure, Art. 1017.
251
Arons & van Boom 2010 (n 228), p. 869.
252
However, in case the settlement funds are not enough to cover the claims that were made (e.g. as a result of
inaccurate estimations), then the outstanding claims would have to be reduced on a pro rata basis; Dutch Civil
Code, Art. 7:909(5).
253
Dutch Civil Code, Art. 7:907(4).
76
elements of the settlement, however, it allows for more flexible solutions and party autonomy
than the UK review.
Different from the collective settlement procedures discussed above, the fairness of voluntary
compensation would not necessarily be scrutinized by a public authority or a court in case the
offer takes places outside of those frameworks. In such instances, the voluntary compensation
efforts could take the form of a contractual agreement with a group of explicitly consenting
victims, akin to an opt-in settlement. However, if such a compensation offer is presented with
a view towards receiving a fine reduction, then a public enforcer could nevertheless conduct a
review. The Dutch fining guidelines highlight voluntary compensation as a mitigating factor if
the infringing undertaking ‘provided full compensation’ to the victims, which implies that the
compensation must have already occurred and the enforcer would therefore not be
‘approving’ or reviewing the compensation as such. On the other hand, the UK voluntary
redress schemes can be developed in parallel to the public investigation and would be
subjected to a two-layered external review process. The first layer includes independent
experts, who would have to devise the schemes and address all the relevant aspects, such as
the target beneficiaries and the type of loss and compensation that are recoverable.254 They
have to balance the wishes of the infringing applicant, the interests of the victims and the
anticipated approval considerations of the CMA, and due to the diverse composition of the
expert body, multifaceted interests could receive an outlet. The second layer of review
involves the final approval process by the CMA. However, the CMA is not meant to carry out
a detailed review of the scheme, especially when it would duplicate or undermine the work of
the independent experts. It assesses the compensation amount offered, the duration of the
scheme, various formal requirements, the position of vulnerable consumers, the objectivity
and independence of the experts, as well as any dissenting views from the infringing
undertaking or the independent experts.255 Overall, although the external review of voluntary
redress schemes is multilayered, a lot of responsibility lies with the independent experts.
(5) Role of the public enforcer
254
CMA Guidance (n 199), 2.12.
CMA Guidance (n 199), 3.5-3.6. In addition to assessing the fairness of the redress scheme, the CMA would
also determine the size of the fine reduction by considering the size of the fine and the existence of other
reductions (i.e. from leniency and settlements), but it would not take into account the nature and gravity of the
underlying competition infringement; CMA Guidance (n 199), 3.31.
255
77
Unlike other UK sector-specific regulation, competition enforcement has not been developed
in accordance with the principles underlying responsive regulation nor has the public enforcer
received redress powers.256 In general, the public enforcement of competition law continues to
be led by command-and-control and deterrence-oriented strategies, which do not view the
promotion of compensation as a priority of the public enforcer. The new regime instituted by
the Consumer Rights Act 2015 altered this paradigm with the introduction of enforcerapproved voluntary redress schemes and thereby integrated the public enforcer into the
promotion of collective ADR. However, it would be premature to note the emergence of a
responsive, compliance-oriented or compensatory enforcement strategy. Similar points can be
made for competition enforcement in the Netherlands. Although compensation efforts are
explicitly taken into consideration as mitigating circumstances when setting the fine,
providing redress is generally not viewed as a priority of the public enforcer. However, in the
Dutch Government’s response to the Green Paper on Consumer Collective Redress, it called
for further investigation regarding the possible role of the public supervisory authority in
contributing to collective redress 'within the framework of executing its public task'. In
particular, although it affirmed the maintenance of the separation between public and private
enforcement, it proposed the exploration of the public enforcer 'promoting the initiative’ of
the infringer to offer voluntary compensation.257 This could demonstrate the openness of the
Netherlands to consider further integration of the public enforcer in the promotion of
collective ADR. Overall, despite the limited connections between compensation and public
enforcement, both the UK and the Netherlands grant an important power to the public
enforcer, namely the inducement of collective ADR through fine reductions.
Conclusion
At first sight, the Netherlands and the UK share many similarities, especially regarding their
enthusiasm to make collective redress and ADR devices more easily available for the victims
and infringing undertakings. However, there are a number of important differences between
their national architectures and the specific collective ADR devices they chose to develop.
The comparative analysis studied the key issues that are crucial for the success of a collective
ADR model under both national scenarios. In addition to outlining the detailed outcomes
regarding the incentives, funding, procedure, fairness guarantees and interplay with public
256
257
Hodges 2015 (n 41), p. 420.
Dutch Government 2008 (n 241), p. 13-14.
78
enforcers, the analysis revealed broader tendencies that distinguish both architectures.
Importantly, the Netherlands reflected a more flexible and functional approach to collective
ADR, whereas the UK opted for a more regulated framework to ensure fairness guarantees.
This chapter analyzed the implications of these policy choices in the abstract, however, the
best lessons to be learned for the development of a European collective ADR model could be
revealed once both national regimes are faced with a competition claim involving large-scale
low-value damage. Presently, neither Member State has had the opportunity to utilize their
regulatory enforcement architecture in this context, but it is undoubtedly important to follow
these developments closely in the coming years and the structured analysis developed in this
section could help frame the discussions that follow.
79
80
Chapter 5: The implications of integrating collective ADR into EU competition
enforcement
The final chapter will be used to reflect on the proposal of integrating collective ADR from a
normative, legal and practical perspective in the context of EU competition law. The first
section will highlight the transformative impact of collective ADR on competition
enforcement and its ability to achieve the goals of deterrence and compensation. The second
section will take a closer look at the Commission’s role as a public enforcer in the architecture
envisioned and its capacity to induce collective ADR, while highlighting the changes that
would have to be made to the current architecture in order to successfully integrate collective
ADR.
5.1 Impact on the public-private interplay and the goals of enforcement
Introducing a regulatory enforcement architecture that facilitates and induces collective ADR
as the first choice compensatory avenue, would undoubtedly have an impact on the current
state of public and private enforcement on the EU and national level. The following
discussions will attempt to shed light on its impact from two angles, namely by addressing the
possible repercussions for the public-private interplay and the ability of collective ADR to
meet the declared goals of competition enforcement.
Public-private transformations
Despite the ease of classifying collective ADR as a ‘particularly private’ device of ‘private
enforcement’, the integration of collective ADR could contribute to transformations that
extend far beyond dispute resolution. The transformative power of collective ADR stems from
public-private tensions between: (1) the public and private enforcement of competition law,
and (2) the public and private devices of private enforcement.
Within the first tension, the public and private enforcement of EU competition law have
traditionally been viewed as serving different regulatory goals, whereby the public enforcer
produces deterrence through fines and private actors deliver compensation through litigation.
While it has been commonly recognized that private enforcement is also capable of producing
81
deterrent effects, it has been less accepted that public enforcement should be producing
compensation. The successful integration of collective ADR into the competition enforcement
architecture would entail some degree of fusion between public and private enforcement in
order to ensure compensation for all victims. While the public enforcer would not be directly
in charge of ordering or dispensing compensation, it would induce voluntary compensation
efforts by taking them into account when determining the size of the fine.258 As a carrot, it
would consider compensation as a mitigating factor, and as a stick, it could consider abusive
or intentionally misleading use of collective ADR as an aggravating factor. Involving the
public enforcer in this manner would not constitute a stark departure from the traditional
distinction between public and private enforcement, however, it could signal the emergence of
a hybrid model of enforcement, which begins to expand the role of competition authorities
beyond policing anti-competitive conduct towards rectifying the broader societal impact of
the infringements. It could also lead to a gradual reassessment of the public enforcement
toolbox. In particular, the facilitation of responsive, behaviourally-informed and complianceoriented strategies for private enforcement could influence the strategies deployed for public
enforcement, possibly leading to a fuller understanding of the reasons behind infringements
and to the promotion of additional enforcement devices, such as self-regulatory compliance
programmes and criminal penalties.
Within the second tension, the public device of private enforcement is traditionally litigation
through state-powered courts, whereas the private device could constitute various out-of-court
dispute resolution processes and non-judicial entities. Introducing collective ADR to the
private enforcement toolbox would reflect the increasing popularity of out-of-court dispute
resolution and the continued exploration of collective devices in the EU. Importantly, these
changes to private enforcement come at a cost, which could affect the prevailing
understanding of how and to what extent justice (or more narrowly, compensation) ought to
be delivered. For instance, with any form of collective proceedings, there are likely to be
significant tradeoffs between individual and collective rights. Similarly, with consensual
proceedings, the agreed outcome is likely to require some concessions from all parties.
Ultimately, the careful integration of collective ADR into a regulatory enforcement
architecture would provide the opportunity to shape these tradeoffs ex ante, ensure minimum
258
For a fuller spectrum of possible public enforcer involvement, see C. Hodges, ‘Mass Collective Redress:
Consumer ADR and Regulatory Techniques’ (2015) European Review of Private Law Vol. 23(5), 829, p. 841845.
82
safeguards and regulate the competition between different devices, entities and jurisdictional
venues.
Building on the transformative themes explored above, an attempt will be made to
conceptualize the contribution of collective ADR towards the goals of competition
enforcement. The two enforcement goals that have been highlighted repeatedly in this thesis
are compensation and deterrence. Firstly, compensation has been viewed as the main goal of
the private enforcement of competition law in Europe.259 The emphasis on an individual right
to full compensation reflects the concept of corrective justice, whereby the aim of the tort
damages claim is for the infringer to remedy the individual harm suffered by the victim.
Secondly, deterrence has been viewed as the main goal of public enforcement and it has also
been accepted as an additional goal, or at least as a possible side-effect, of private
enforcement. In the EU competition law context, the acknowledgement of the regulatory
dimension of private claims has largely been influenced by the significant deterrent role
assigned to private enforcement in the US. However, with both public and private
enforcement, the concept of ‘optimal deterrence’ is likely to remain elusive for a variety of
reasons, including the lack of direct coordination between the two. With these considerations
in mind, the following sections will take a closer look at the impact of collective ADR on
achieving deterrence and compensation.
The impact on deterrence
In the EU, deterrence is predominantly produced through imposing financial penalties on the
infringers, which can be derived from both public and private enforcement. From a
behavioural perspective, the precise deterrent effects of enforcement are difficult to measure,
since it would require some insight into the subjective perceptions of the infringing
undertakings. Nevertheless, the fundamental deterrence capacity of financial penalties has
259
For instance, the Commission’s press release accompanying the Damages Directive stated the following:
“Contrary to the US system, the proposal does not seek to leave the punishment and deterrence to private
litigation. Rather, its main objective is to facilitate full and fair compensation for victims once a public authority
has found and sanctioned an infringement.” The emphasis on follow-on claims is notable. More broadly, this
view was also presented in the 2008 White Paper on Damages Actions, p. 3: “The primary objective of this
White Paper is to improve the legal conditions for victims to exercise their right under the Treaty to reparation of
all damage suffered as a result of a breach of the EC antitrust rules. Full compensation is, therefore, the first and
foremost guiding principle.” [emphasis added]
83
been subject to criticism.260 To illustrate, the combination of capping the size of the financial
penalty at 10% of the annual turnover, the 13% estimated detection rate for cartels and the
25% estimated average overcharge would generally make infringements the economically
rational choice for undertakings.261 The low likelihood of subsequent private damages claims,
combined with the existence of legal insurance, would not increase the financial penalty
significantly. Moreover, in order to achieve ‘optimal deterrence’, the combined financial
penalty would often have to be enormous. Even if such an outcome would be achieved, it
could result in insolvency for many infringing undertakings, which could in turn have several
undesirable effects. Moreover, financial penalties would rarely have an effect on the
individuals that were directly responsible for the infringements.262 Understanding the limits of
financial penalties enables a more informed assessment of collective ADR’s impact on
deterrence. The list of negative and positive effects compiled below is not exhaustive.
Negative effects: (i) In case collective ADR results in a fine reduction, the level of deterrence
derived from the financial penalty would seemingly be reduced, since a portion would be
subtracted from the financial penalty. A fine reduction would essentially convert some
deterrence into compensation by redistributing some of the illegal profits directly among the
victims of the violation. However, since the exact size of the anticipated fine reduction might
not always be known in advance and it might also be uncertain whether all of the relevant
victims end up claiming their share of the compensation, then the total sum paid by the
infringer would not necessarily be foreseeable. In sum, deterrence would only be reduced in
case the compensation amount is indeed smaller than the fine reduction received. (ii) If
collective ADR produces fewer costs for the infringing undertaking than litigation, then its
use would directly limit the size of the financial penalty. This reduction of deterrence is
inevitable in cases where the threat of litigation is used to induce collective ADR, since such
savings would be one of the key reasons why the infringers agreed to it. However, an
exceptional scenario could be envisioned if the infringer engaged in collective ADR under the
mistaken belief that the litigation claim would succeed, when in practice it would have not.
260
W. Wils, The Optimal Enforcement of EC Antitrust Law (Kluwer Law International 2002), ch. 8; D.A. Crane,
'Optimizing Private Antitrust Enforcement' (2010) Vanderbilt Law Review Vol. 63(3), 673.
261
A. Renda, ‘Private Antitrust Damages Actions in the EU’ in H.-W. Micklitz & A. Wechsler (eds), The
Transformation of Enforcement: European Economic Law in Global Perspective (Hart Publishing 2016), p. 275276.
262
Wils 2002 (n 260).
84
Positive effects: (i) If collective ADR took the form of voluntary compensation in return for
fine reductions, then such ‘negotiated’ infringement decisions might be less likely to be
appealed.263 In addition, competitors that are reputationally sensitive might be less likely to
challenge an infringement decision that involved compensating victims. This, in turn, would
save resources for the public enforcer, who could invest more in other investigations and
thereby the overall deterrence effect of public enforcement could be increased. (ii) If
collective ADR leads to an increase in compensatory outcomes, then competition law would
become more visible to the general public. This could make future victims more aware of
their rights and more likely to engage in private enforcement or report infringements to the
public enforcer.264 At first sight, the confidentiality of certain ADR proceedings would limit
such publicity, however, the extensive notification requirements for collective proceedings
that include a large number of non-participating victims could nevertheless make the
existence of compensatory activities sufficiently public. Moreover, once infringers have
agreed to compensate, they could additionally reap reputational benefits by ‘advertising’ their
corporate responsibility and voluntary compensation efforts. As a side effect, the increased
publicity for competition law among other market participants could have a positive effect on
the deterrence of infringements.
The impact on compensation
Subject to the principles of effectiveness and equivalence, Courage and Manfredi established
that ‘full compensation’ should be made available to everyone who suffered a loss resulting
from a competition infringement.265 Full compensation includes the actual loss as well as lost
profits and interest, which would accrue from the time the harm occurred until the
compensation is paid.266 Full compensation is meant to represent the optimal level and any
form of overcompensation through punitive or multiple damages has been expressly
rejected.267 However, at the premise of this thesis lies the concern that the current state of
private enforcement is rather producing undercompensation, particularly for victims suffering
from large-scale low-value harm. Moreover, the precise concept of full compensation is
263
J.H.J. Bourgeois & S. Strievi, ‘EU Competition Remedies in Consumer Cases: Thinking Out of the Shopping
Bag’ (2010) World Competition Vol. 33(2), 241, p. 254.
264
Bourgeois & Strievi 2010 (n 263), p. 253-254.
265
Courage Ltd v Crehan (n 26); Manfredi and others (n 26). These principles are now enshrined in Art. 3 and 4
of the Damages Directive (n 2).
266
Damages Directive (n 2), Recital 12.
267
Damages Directive (n 2), Art. 3 (3).
85
difficult to achieve in practice, since the quantification of harm and the determination of the
final compensatory amount are likely to be difficult and require approximations. In addition,
the actual compensatory sum received by the victims is likely to be smaller than the full
compensation amount once the costs of bringing the claim are taken into account.
Understanding the limits of traditional damages claims enables a more informed assessment
of collective ADR’s impact on compensation. The list of negative and positive effects
compiled below is not exhaustive.
Negative effects: (i) If the victims could have received a higher amount of compensation
through litigation, then collective ADR would directly lessen the amount of compensation
paid. This tradeoff would occur in most cases, where the threat of litigation is used to induce
collective ADR, since such savings would be one of the key reasons why the infringers agreed
to it. (ii) The scope of the compensation achieved through collective ADR might not cover the
precise losses prescribed by the concept of full compensation. However, it is questionable
whether collective litigation could achieve that standard either, considering the level of
approximation involved, for instance, with the quantification of harm. (iii) Similarly, the
scope of the compensation achieved through collective ADR might limit the categories of
victims that are eligible for compensation. Selective compensating could result in weakening
the position of those victims that would like to continue with collective litigation.
Positive effects: (i) The total costs of collective ADR could be lower than the total costs of
litigation, even if litigation would have provided a larger compensatory sum. The total cost of
the claim includes the compensation amount, but also the legal fees and other costs associated
with prolonged representation of the claim. Moreover, the total costs for society are likely to
be even greater if the costs of using the court system are considered. In addition, the total
costs would be even higher if the costs borne by the infringing undertaking would ultimately
be passed down to other parties like consumers. (ii) A consensual agreement reached by the
parties though collective ADR is less likely to pose problems in the enforcement stage, since
it would generally be self-enforcing due to consensus. An adversarial outcome reached
through litigation could entail additional costs in case of appeal or other forms of contestation.
Furthermore, with collective ADR, the infringing undertaking would agree to a compensation
amount that it could actually deliver, while litigation could result in a sum that might lead to
insolvency, which could leave the victims without any compensation. (iii) The flexibility of
collective ADR could help overcome procedural restrictions that could make collective
86
litigation problematic, for instance, the determination of a ‘common interest’ among victims
with conflicting claims (direct and indirect purchasers), the quantification of harm and the
evidence that must be provided for victims to claim their share. These procedural restrictions
would have a direct impact on whether full compensation would be achieved.
Deterrence, compensation and ‘resolution’
As demonstrated by the analysis above, the integration of collective ADR into the regulatory
enforcement architecture would certainly have an impact on achieving deterrence and
compensation. The particular outcome reached through collective ADR is unlikely to
constitute ‘optimal deterrence’ or ‘full compensation’. However, instead of understanding that
outcome as suboptimal deterrence and incomplete compensation, it could be defined as a
distinct concept. In many cases, collective ADR would result in the most just and efficient
outcome that can be achieved, while taking into account all relevant circumstances and the
interests of all affected parties - the goal achieved in this manner could be conceptualized as
‘resolution’.
In essence, resolution represents a combination of what is just and efficient, while taking into
account the substantive, procedural and social context and balancing the interest of all parties
involved - not just the victims and infringers, but also the competition authorities, courts and
society as a whole. Ultimately, by following these principles, resolution attempts to remedy
the negative consequences of an infringement insofar as possible. Since one of the negative
consequences of an infringement is the economic harm suffered by individual victims, then
compensation could itself be viewed as one facet of achieving resolution. However, resolution
should be understood as going further and encompassing broader restorative effects for all
parties that were affected by the infringement. Put differently, resolution goes beyond
corrective justice and places more emphasis on distributional justice, market balance and
notions of collective fairness. Resolution is not confined to remedying the visible
consequences of past infringements - it could also introduce a future-oriented dimension,
whereby genuine attempts to proactively resolve both the consequences as well as the causes
of infringements, would become more rewarding for the infringers. Finally, since no single
enforcement strategy has been proven to be perfectly optimal or capable of addressing all of
the negative consequences of competition infringements, then resolution could signify a third
87
strategy that not only fills the gaps left by deterrence and compensation, but also provides for
coordination between the combined effects of all enforcement actions taken.
The concept could be expanded even further and applied more broadly in the context of
public enforcement, but for the purposes of the present inquiry, the emphasis will be on
collective ADR’s ability to achieve ‘resolution’. From the outset, adding a speedier, cheaper
and more consensual private enforcement avenue that is capable of covering as many victims
as possible is very likely to be welfare-enhancing and restorative. In most cases, the outcome
reached through collective ADR would be the most efficient way achieving compensation, as
far as private enforcement devices are concerned. Moreover, with collective ADR, an
immediate and adversarial interaction between the infringers and victims would be replaced
with a more mediated interaction that involves consensus-building and distributive
considerations. The lack of vindication could have a positive impact on the emergence of
cooperative and compliance-oriented attitudes in the infringing undertaking. The achievement
of a just outcome for the affected parties is further legitimized by consent and emerges from
the consideration of all alternative avenues that are available. In case the outcome binds nonparticipating victims, then their consent deficiencies could be remedied by a public fairness
review. It is submitted that perfect procedural justice is unlikely to be obtained in a world of
incomplete information, interpretable legal norms and private interests, neither by collective
ADR nor collective litigation. However, collective ADR ultimately satisfies the notion of
access to justice that is built on a contextual understanding of the needs, solutions and impacts
of the legal response, taking into account the cultural, economic and psychological realities of
individual and organizational behaviour.268 Furthermore, collective ADR can advance interests
that go beyond the restoration of individual or aggregated rights. Put differently, the
consensual resolution of the dispute could produce positive externalities that transcend the
damages claim. For instance, the courts would be less burdened and the taxpayers would be
spared from funding litigation that could be resolved out of court. Moreover, the leftover
funds could be directed to a cy-prés outlet that serves broader societal interests. Leftover
funds from compensation received in exchange for a fine reduction would be particularly
well-suited for serving such interests, since it was deducted from a financial penalty that
would have otherwise been directed to a public source.269 The use of collective ADR in return
268
This multifaceted understanding of ‘access to justice’ was acknowledged by Cappelletti 1993 (n 128), p. 283.
For instance, the compensation could be viewed as serving both past and future consumers, Bourgeois &
Strievi 2010 (n 263), p. 244.
269
88
for fine reductions would also ensure some coordination between the effects of public and
private enforcement, since a reassessment of the fine would occur in light of the
compensatory sum paid. In addition, collective ADR could serve a gap-filling purpose, not
only in terms of providing compensation for the victims that would otherwise be left without
a remedy, but also by creating an outlet for compliance behaviour. When compensation efforts
are taken into account by the public enforcer or the courts, a connection is created between
liability and self-regulation, which could make the infringing undertakings reflect internally
on the causes for infringements and ensure that adequate prevention and mitigation structures
are in place.270 The compliance-oriented rationale behind inducing the use of collective ADR
is not always easy to reconcile with the deterrence-oriented rationale that is dominant in
competition enforcement. However, when regulating behaviour, perhaps the underlying
insights should in fact come from behavioural research, not from economic theory. If the
current strategies fall short of achieving optimal outcomes, then the exploration of new and
increasingly behaviourally-informed public and private enforcement strategies should be
encouraged. Ultimately, the introduction of collective ADR into the regulatory enforcement
architecture serves as just one example of how the ‘resolution’ approach could work, but it
can nevertheless be viewed as an important first step in the right direction.
5.2 Collective ADR in the Commission’s public enforcement policy
In the regulatory enforcement architecture envisioned, the public enforcer would have an
important role in inducing collective ADR. Although the initiation of collective ADR would
not always be contingent on the participation of the public enforcer, their early involvement
would lead to the speediest form of ‘resolution’ in many cases. The European Commission’s
broader policy stance on collective redress, ADR and private enforcement has been discussed
in the first chapter and will not be recapitulated here at length. As suggested by the research
question in this thesis, collective ADR has not yet been integrated into the competition
enforcement architecture on the EU level. This section will now go into more depth about the
Commission’s current approach to collective ADR as the public enforcer of EU competition
law. The first part will address the subject of inducing collective ADR through fine
reductions, after which the focus will shift to other elements that can indicate the
Commission’s capacity to integrate collective ADR and facilitate a more compliance-oriented
enforcement strategy, namely the availability of special negotiation procedures and the
270
Parker 2002 (n 36), p. 253.
89
treatment of self-regulatory compliance efforts. Each part will conclude with an assessment
on whether any changes need to be introduced in order to successfully integrate collective
ADR.
Treatment of voluntary compensation
As discussed earlier, the Commission has predominantly viewed public and private
enforcement as separate disciplines, serving the different goals of deterrence and
compensation. It is therefore unsurprising that the official position holds that no form of
compensation provided to the victims of competition infringements would affect the public
enforcement process and the Commission’s Fining Guidelines do not list compensation as a
mitigating circumstance.271 However, at least two objections could be made in this regard:
Firstly, the Damages Directive offers a different perspective on whether compensation should
affect public enforcement. To begin with, recital 5 lists ‘public enforcement decisions that
give parties an incentive to provide compensation’ as an example of an ‘alternative’ private
enforcement mechanism. The concept of ‘public enforcement decisions’ is only mentioned in
this particular recital and not elaborated upon elsewhere in the Directive using that
terminology. However, this fusion of public and private enforcement becomes most evident in
Art. 18 (3), which explicitly allows competition authorities to take compensation into account
as a ‘mitigating factor’ when setting the fine. This provision strengthens the legitimacy of
inducing collective ADR through fine reductions, even if the Commission’s own guidelines
do not yet provide for this possibility. The Commission’s proposal that the NCAs should
consider offering fine reductions in exchange for compensation is, at the very least,
illustrative of the fact that it does not hold any fundamental objections to the idea.
Secondly, the Commission’s own case law includes several instances of taking voluntary
compensation into account when determining the size of the financial penalty.272 The earliest
instance occurred in General Motors Continental, where the Commission considered
compensation as one of several mitigating circumstances. There were five cases of excessive
charging that amounted to an abuse of dominance and GMC offered to voluntarily reimburse
271
Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003, OJ
C 210.
272
Ioannidou 2015 (n 4), p. 172-174.
90
the overcharge in three cases and also paid a compensation amount that exceeded the
overcharge in two cases. 273 In Pre-Insulated Pipe Cartel, the Commission expressly
considered compensation as the only mitigating circumstance for one member of the cartel.
The cartelist ABB reached an agreement for a 'substantial sum' with Powerpipe, who was a
victim and a direct competitor, and copy of the confidential settlement was sent to the
Commission. 274 In Nintendo, the Commission expressly considered compensation as a
mitigating circumstance and calculated a fine reduction of 300 000 euros in light of the
compensatory amount. Nintendo offered 'substantial' financial compensation to 11 victims, 9
of which accepted the offer. 275 When assessing these three cases, it is notable that they all
involve out-of-court consensual compensation, but not all of them feature a significant
collective dimension. The compensation was offered to direct purchasers and competitors,
who are less likely to suffer the type of large-scale low-value harm that indirect purchasers
and consumers experience. Nevertheless, it is clear that the Commission has in fact
encouraged voluntary compensation and offered fine reductions, subject to its discretion. The
discretionary nature was confirmed in Archer Daniels Midland, where the General Court
stated that the Commission is not obligated to take compensation into account as a mitigating
circumstance just because it has chosen to do so in previous cases.276
Against this background and the experience of the two national systems analysed in chapter 4,
in order to integrate collective ADR into the regulatory enforcement architecture on the EU
level, the Fining Guidelines should be amended to include compensation as a mitigating
circumstance. When assessing whether the compensation offered qualifies for a fine
reduction, the Commission could make use of the benchmarks of the fairness review
envisioned in chapter 3, which included scrutinizing the compensation achieved, the
notification and participation rights of non-participating victims, the role of the representative
entity and the plans for unclaimed funds. If compensation is considered as a mitigating
circumstance, then the substantive assessment and the fine reduction amount would largely be
subject to the Commission’s discretion and work on a case-by-case basis. The issuance of
guidelines or best practices for the infringing undertakings could help improve the
273
Commission Decision of 19 December 1974 General Motors Continental (IV/28.851) [1975] OJ L29/14,
para. 18.
274
Commission Decision of 21 October 1998 Pre-Insulated Pipe Cartel (IV/35.691/E-4) [1999] OJ L24/1, para.
127.
275
Commission Decision of 30 October 2002 PO Video Games (COMP/35.587), PO Nintendo Distribution
(COMP/35.706), Omega - Nintendo (COMP/36.321) [2003] OJ L 255, para. 440-441.
276
Case T-59/02 Archer Daniels Midland Co v Commission [2006] ECR II-3627, para. 353-355.
91
predictability of the process and increase the attractiveness of collective ADR. It should be
noted that the inclusion of compensation in exchange for fine reductions could have
implications that go beyond inducing collective ADR. With a general power to consider
compensation as a mitigating circumstance, the public enforcer could take any form of
compensation into account, including the sums paid in the context of adversarial collective
litigation. While the encouragement of voluntary and consensual compensation is the desired
outcome, fine reductions should not necessarily be exclusively limited to ADR outcomes. In
addition to the general inclusion of compensation as a mitigating circumstance, a more
regulated framework could be considered. As seen in the UK architecture, the CMA was
tasked with the approval of voluntary redress schemes, which would have to be designed in
accordance with specific guidelines and could result in up to a 20% fine reduction. In contrast
to the mitigating circumstance scenario highlighted above, the fine reduction would be
directly contingent on the voluntary and consensual nature of the compensation, among other
requirements. A more regulated framework could include requirements that go beyond the
basic considerations in the fairness review and provide increased predictability concerning the
steps that would have to be taken in order to receive a fine reduction. The regulated
framework could also envision a role for external experts. For instance, the voluntary redress
schemes in the UK prescribed the appointment of neutral third parties, who would largely be
in charge of devising the schemes. Some of the public enforcer’s assessment tasks could
thereby be delegated to third parties. On the one hand, this would save public resources and
allow the enforcer to focus on the ongoing investigation, on the other, the diminished public
review and the involvement of third parties could readily create some risks, which would have
to be addressed. Ultimately, the integration of collective ADR into the regulatory enforcement
architecture in exchange for fine reductions requires the acceptance of compensation as a
mitigating circumstance and if, in addition, a particular collective ADR process is deemed to
be desirable, then it could be promoted with specific guidelines.
Availability of negotiation procedures
Special negotiation procedures could serve as an additional venue for voluntary compensation
incentives to emerge. The Commission’s toolbox includes two procedures that can be relevant
in this regard: cartel case settlements and commitment decisions.
92
Firstly, Art. 10a of Regulation 773/2004 and the Settlements Notice established a negotiated
settlement procedure in cartel cases, which offers the cartelists a chance to admit liability,
cooperate and receive a 10% fine reduction, subject to the Commission’s discretion. Cartel
case settlements are essentially procedural efficiency devices that are used to speed up cartel
investigations, which will result in an infringement decision. The connection to inducing
collective ADR is not immediately visible, since the settlement submission itself does not
include any direct considerations for compensation. However, as demonstrated by the cases
that emerged in the Netherlands and the UK, the settlement procedure and the early
admittance of the infringement could make the cartelists consider the consequences of
subsequent private enforcement sooner and take proactive compensatory action to mitigate
risks.
Secondly, Art. 9 of Regulation 1/2003 established the commitment decision procedure, which
follows a preliminary assessment and offers the infringing undertakings a chance to accept
legally binding commitments in order to have the investigation terminated, subject to the
Commission’s discretion. The commitment decision is silent on whether there was an
infringement, which would allow the NCAs and national courts to nevertheless find an
infringement and award damages. The procedure is not used in hardcore cartel cases, where
the only appropriate remedy would be a financial penalty,277 and since large-scale low-value
harm often occurs as a result of such infringements, then this procedure would have limited
relevance in such instances. The Commission has explicitly approved of a compensatory
commitment in Deutsche Bahn, which concerned the pricing practices that DB Energie
applied to German railway undertakings that bought electricity for powering locomotives.278
DB Energie offered to pay a one-time compensatory sum to all German railway undertakings
that did not belong to the DB Group. The sum consisted of 4% from their yearly invoice,
based on the latest annual pricing conditions that were in place, and the Commission deemed
it to be an adequate means for countering margin squeeze effects.279 Interestingly, in response
to comments that the compensatory sum should have been even greater, the Commission
stated that this compensatory payment should not be equated with ‘compensation for harm
suffered through possible anti-competitive behaviour’ because commitment decisions do not
result in a finding of infringement and therefore this compensatory payment is not ‘aimed to
R. Whish & D. Bailey, Competition Law (8th edn, Oxford University Press, 2015), p. 274.
Commission Decision of 18 December 2013 Deutsche Bahn (COMP/AT.39678/Deutsche Bahn I,
COMP/AT.39731/Deutsche Bahn II).
279
Deutsche Bahn (n 278), para. 91-92.
277
278
93
directly compensate harm suffered’ as a result of an infringement. 280 It is questionable
whether this is in fact the case, especially with regards to any subsequent damages claims
following a possible finding of infringement by the NCAs or national courts. If the infringer
offers voluntary compensation, then the victim, by explicitly accepting the sum, would have
at least a part of their harm restored for tort law purposes. Moreover, the extent of the
coverage could be subject to the terms of any agreement reached between the infringer and
the victim. Therefore, the compensation achieved through commitments could have an effect
on the compensation achievable for the harm resulting from an infringement.
Finally, outside of the formal framework of the two negotiation procedures discussed above,
the Commission has informally facilitated voluntary compensation in several cases, although
the connection between the incentives provided and the voluntary compensation was less
direct.281 To conclude, special negotiation procedures can be a relevant venue for inducing
collective ADR, however, the effects of the procedure’s outcome should be taken into account
when encouraging compensation at that stage, particularly if the outcome is the closing of the
investigation. For scenarios that involve large-scale low-value harm, fine reductions would
generally be a more appropriate venue for inducing collective ADR. Ultimately, the decision
whether to take voluntary compensation into account in the context of such negotiation
procedures could remain at the discretion of the Commission and be decided on a case-bycase basis.
Treatment of self-regulatory compliance
The promotion of self-regulatory compliance efforts can be indicative of the openness to
facilitate compliance-oriented enforcement strategies, not only for public, but also for private
enforcement. Despite the Commission ‘welcoming and supporting’ all compliance efforts by
undertakings, its policy stance has been rather categorical in excluding the direct relevance of
280
Deutsche Bahn (n 278), para. 93.
For example, in Macron, the Commission closed the investigation after the dominant undertaking made
informal commitments, which included paying a compensatory sum to a single complainant; in Rover, the
Commission did not initiate an investigation, after Rover notified a revised discount scheme and offered a 1 000
000 GBP donation to Which?, the largest consumer association in the UK; in Sony-Philips, the Commission
cleared a Standard Licensing Agreement, which included a commitment by Philips to pay each holder of a
license 10 000 USD, amounting to a total of 800 000 USD; Ioannidou 2015 (n 4), p. 173; Bourgeois & Strievi
2010 (n 263), p. 248.
281
94
compliance programmes for the public enforcement282 and the Fining Guidelines do not list
such programmes or other compliance efforts as relevant mitigating circumstances. The
Commission has been confronted with the existence of self-regulatory compliance
programmes on several occasions, however, it has never granted a fine reduction as a result of
a pre-existing compliance programme. In some early cases, the Commission did take
compliance efforts into account, but only when the compliance programme was set up
following the investigation. 283 More recently in Archer Daniels Midland, in addition to
offering voluntary compensation to the victims, ADM featured a 'rigorous and ongoing' selfregulatory compliance programme and its level of commitment indicated a genuine desire to
avoid infringements, however, the Commission did not consider it to be relevant. In the
appeal, the General Court ultimately established that the Commission is not obligated to take
these efforts into account.284
It is therefore clear that compliance efforts are currently not integrated into the public
enforcement process. However, it is submitted that the treatment of self-regulatory
compliance programmes should be considered in the broader context of the limited deterrence
effects that can be achieved through financial penalties. The encouragement of compliance
could be relevant not only for public enforcement,285 but also for private enforcement and the
inducement of collective ADR. On a more practical level, a well-functioning compliance
programme could be utilized not only for the detection of infringements, but also for the
mitigation of negative consequences, for instance, by tasking the same staff with the
development and management of voluntary redress schemes. For the undertakings that are
reputationally sensitive, both compliance and compensation would signal a similar
willingness to accept responsibility and could be presented together as a complimentary
package that mitigates past infringements and prevents future infringements. Moreover,
including an ex ante commitment to mitigate harm within the self-regulatory compliance
282
European Commission, Compliance matters: What companies can do better to respect EU competition rules,
Luxembourg: Publications Office of the European Union, 2012, p. 21: “If a company which has put a
compliance programme in place is nevertheless found to have committed an infringement of EU competition
rules, the question of whether there is any positive impact on the level of fines frequently arises. The answer is:
No.“
283
W. Wils, ‘Antitrust Compliance Programmes and Optimal Antitrust Enforcement’ (2013) Journal of Antitrust
Enforcement, Vol. 1(1), 52, p. 53; the fine reductions were granted in certain decisions taken between 19821992.
284
Archer Daniels Midland (n 276), para. 356-359.
285
For a discussion on the potential of compliance programmes for public enforcement, see Wils 2013 (n 283);
Lachnit 2014 (n 171); Ottow 2015 (n 170), p. 183-185; D. Geradin, 'Antitrust Compliance Programmes and
Optimal Antitrust Enforcement: a Reply to Wouter Wils' (2013) Journal of Antitrust Enforcement Vol. 1(2), 325.
95
programme could help the public enforcer in their assessment of the compliance efforts, since
such a promise could be an indicator that the compliance programmes were not meant to be
merely ‘window dressing’. In such cases, the undertakings would not necessarily have to be
held accountable for a failure to keep their promise, however, their subsequent refusal to even
accommodate collective ADR negotiations could be viewed as an aggravating circumstance
and provide an opportunity to impose higher fines. Finally, and perhaps most importantly, the
promotion of compliance behaviour might contribute to the creation of a genuine internal
compliance culture that also involves compensating victims in case of failure. In sum, the
treatment of self-regulatory compliance programmes could be viewed as one component of
the broader exploration of untapped compliance-oriented strategies and it might have a
positive impact on inducing voluntary compensation. However, collective ADR is not
ultimately contingent on the facilitation of self-regulatory compliance programmes.
96
Conclusion
This thesis proposes that collective ADR has the potential to provide just and efficient
resolution for all parties directly affected by a competition infringement. In order to achieve
that goal, collective ADR should act as the first choice redress avenue in a regulatory
enforcement architecture that assigns roles for both private and public actors. The successful
inducement of collective ADR necessitates a responsive strategy that takes into account the
intricacies of behavioural research and the multi-layered incentives that need to be provided
for the infringers and their victims. The public enforcer should be assigned the role of
inducing voluntary compensation efforts through fine reductions, which necessitates some
level of fusion between public and private enforcement. The key concern with collective ADR
was deemed to be the existence of genuine consent to the proceedings and its binding
outcome, particularly in case a large amount of non-participating parties would be involved.
The use of an opt-out model was found to be most desirable for ensuring the broadest possible
coverage of victims and for increasing the attractiveness of a once-and-for-all global solution
for the infringing undertakings. The increasing acceptance of opt-out models in European
civil justice systems is illustrated by the architectures of both the UK and the Netherlands.
The due process and fairness concerns should be scrutinized in an ex post fairness review that
gives due consideration to the tradeoffs between individual and collective interests that made
a consensual outcome possible in the first place. The extent to which collective ADR is
‘regulated’ would have a direct impact on its functioning. The fairness review could be
conducted by courts, as seen with the settlement regimes in the Netherlands and the UK, or by
the public enforcer with possible assistance from neutral third parties, as seen with the
voluntary redress schemes in the UK.
Furthermore, it is submitted that introducing collective ADR would not be in conflict with
existing legislation - to the contrary, it would build upon the private enforcement provisions
that have already been introduced. The Damages Directive defined ‘consensual dispute
resolution’ as ‘any mechanism enabling parties to reach the out-of-court resolution of a
dispute concerning a claim for damages’,286 which would cover all forms of collective ADR
envisioned in this thesis. The Directive explicitly encouraged and promoted consensual
dispute resolution processes that would cover ‘as many injured parties and infringers as
286
Damages Directive (n 2), Art. 2 (21).
97
legally possible’, 287 which indicates a preference for collective forms of ADR with the
broadest possible application. The Directive also explicitly encouraged the involvement of
public enforcers and allowed them to take compensation into account as a mitigating
circumstance. 288 In addition, the Directive introduced specific procedural provisions that
would also apply to any future models of collective ADR, namely the suspension of the
limitation period and ongoing litigation proceedings for the duration of the consensual dispute
resolution process, and the clarification of the ‘joint and several liability’ relationship between
settling and non-settling infringers. 289 In light of these existing provisions, the Damages
Directive could serve as a suitable forum for introducing additional collective ADR
provisions, such as the benchmarks of the fairness review. However, for reasons of
consistency, it would be advisable to compile all collective redress provisions in a single
instrument. Should a European collective litigation device apply horizontally across different
sectors, then it would be advisable to include collective ADR provisions in the same
instrument. A fuller assessment of such policy choices would go beyond the scope of this
thesis, however, it is submitted that the lack of a competition-specific legislative instrument
would not hinder the functioning of collective ADR in the regulatory enforcement architecture
envisioned.
The following section will take stock of the changes that would have to be made to the current
architecture in order to successfully integrate collective ADR. To begin with, the development
of the regulatory enforcement architecture envisioned in this thesis would be illustrative of
specific policy positions:
● The facilitation of atypical private enforcement structures in order to ensure
compensation and close the enforcement gap for victims suffering from large-scale
low-value damage.
● The exploration of behaviourally-informed and compliance-oriented enforcement
strategies that promote voluntary compensation as the first choice private enforcement
avenue.
287
Damages Directive (n 2), Recital 48.
Damages Directive (n 2), Art. 18 (3).
289
Damages Directive (n 2), Art. 18 & 19.
288
98
● The recognition of ‘resolution’ as a desirable goal of enforcement and ensuring that
collective ADR achieves that goal by leading to just and efficient compensatory
outcomes for all affected parties.
● The promotion of coordination and cooperation between public and private
enforcement in order to induce collective ADR.
● The acceptance of possible tradeoffs that occur in collective and consensual processes
that include a large number of non-participating parties.
Meeting the objectives above could require the following modifications:
● An amendment to the fining policy of competition infringements, which recognizes
compensation as a mitigating circumstance and provides specific guidelines on the
steps that would have to be taken in order to receive a fine reduction.
● The introduction of a collective redress device that introduces collective litigation and
collective ADR. The device should permit opt-out applications, particularly when
victims suffered from large-scale low-value damage.
● The introduction of a reasonable fairness review to be applied by public enforcers and
courts for collective ADR outcomes, particularly for those involving a large number of
non-participating parties.
● An amendment to the European private international law regime that clarifies and
ensures the effective cross-border recognition and enforcement of collective ADR
outcomes.
To conclude, collective ADR is capable of achieving the effective, efficient and fair private
enforcement of competition law and should be explored further as a policy option on the
European level. Ultimately, this thesis laid out the basic groundwork that could be used to
develop a more detailed assessment of the multifaceted implications of integrating collective
ADR into existing frameworks.
99
100
Bibliography
Primary sources
Legislation
Directive 2014/104/EU of the European Parliament and of the Council of 26 November 2014
on certain rules governing actions for damages under national law for infringements of the
competition law provisions of the Member States and of the European Union, OJ L 349.
Directive 2013/11/EU of the European Parliament and of the Council of 21 May 2013 on
alternative dispute resolution for consumer disputes and amending Regulation (EC) No
2006/2004 and Directive 2009/22/EC (Directive on consumer ADR), OJ L 165.
Directive 2008/52/EC of the European Parliament and of the Council of 21 May 2008 on
certain aspects of mediation in civil and commercial matters, OJ L 136.
Regulation (EC) No 1393/2007 of the European Parliament and of the Council of 13
November 2007 on the service in the Member States of judicial and extrajudicial documents
in civil or commercial matters, OJ L 324.
Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12
December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and
commercial matters (recast), OJ L 351.
Council Directive 93/13/EEC of 5 April 1993 on unfair terms in consumer contracts, OJ L
095.
Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules
on competition laid down in Articles 81 and 82 of the Treaty, OJ L 1.
Commission Regulation (EC) No 773/2004 of 7 April 2004 relating to the conduct of
proceedings by the Commission pursuant to Articles 81 and 82 of the EC Treaty, OJ L 123.
Commission Recommendation of 11 June 2013 on common principles for injunctive and
compensatory collective redress mechanisms in the Member States concerning violations of
rights granted under Union Law, OJ L 201 .
Commission Notice on the conduct of settlement procedures in view of the adoption of
Decisions pursuant to Article 7 and Article 23 of Council Regulation (EC) No 1/2003 in cartel
cases, OJ C 167.
Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation
No 1/2003, OJ C 210.
Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York,
1958).
101
Convention on the Service Abroad of Judicial and Extrajudicial Documents in Civil or
Commercial Matters (Hague, 1965).
US
Federal Rules of Civil Procedure
UK
Civil Procedure Rules
Competition Act 1998
Consumer Rights Act 2015
Competition Appeal Tribunal Rules 2015
NL
Civil Code (Burgerlijk Wetboek)
Code of Civil Procedure (Wetboek van Burgerlijke Rechtsvordering)
Competition Act (Mededingingswet)
Act on the Collective Settlement of Mass Damage Claims (Wet Collectieve Afwikkeling
Massaschade)
Act on the Enforcement
consumentenbescherming)
of
Consumer
Protection
Law
(Wet
handhaving
Case law
European Commission
Commission Decision of 19 December 1974 General Motors Continental (IV/28.851) [1975]
OJ L29/14
Commission Decision of 21 October 1998 Pre-Insulated Pipe Cartel (IV/35.691/E-4) [1999]
OJ L24/1
Commission Decision of 30 October 2002 PO Video Games (COMP/35.587), PO Nintendo
Distribution (COMP/35.706), Omega - Nintendo (COMP/36.321) [2003] OJ L 255
Commission Decision of 9 November 2010 Airfreight (COMP/39258) [2010]
Commission Decision of 18 December 2013 Deutsche Bahn (COMP/AT.39678/Deutsche
Bahn I, COMP/AT.39731/Deutsche Bahn II)
102
Court of Justice of the European Union
Case C-453/99, Courage Ltd v Crehan [2001] ECR I-6297
Joined Cases C-295/04 to C-298/04 Manfredi and others [2006] ECR I-6619
Case T-59/02 Archer Daniels Midland Co v Commission [2006] ECR II-3627
Joined Cases C-317/08-C-320/08 Alassini and others v Telecom Italia [2010] ECR I-2213
Cases T-9/11 Air Canada, T-28/11 Koninklijke Luchtvaart Maatschappij, T-36/11 Japan
Airlines, T-38/11 Cathay Pacific Airways, T-39/11 Cargolux Airlines International, T-40/11
Latam Airlines Group and Others, T-43/11 Singapore Airlines and Others, T-46/11 Deutsche
Lufthansa and Others, T-48/11 British Airways, T-56/11 SAS Cargo Group and Others, T62/11 Air France-KLM, T-63/11 Société Air France and T-67/11 Martinair Holland v
Commission
US
American Express Co. v. Italian Colors Restaurant, 570 U.S. ___ (2013)
AT&T Mobility LLC v. Concepcion, 563 U.S. 333 (2011)
Green Tree Financial Corp. v. Bazzle, 539 U.S. 444 (2003)
Stolt-Nielsen S. A. v. AnimalFeeds Int’l Corp., 559 U.S. 662 (2010)
UK
Emerald Supplies Ltd v British Airways Plc [2009] EWHC 741 Ch, [2010] CH 48
The Consumer Association v JJB Sports Plc (2007) CAT Case No. 1078/7/9/07
UK Independent Schools (2006) OFT Decision No. CA98/05/2006
NL
Superunie vs Interpay, Dutch Competition Authority, 28 April 2004, Case no. 2978/56
DES, Amsterdam Court of Appeal, 1 June 2006, LJN: AX6440
Dexia, Amsterdam Court of Appeal, 25 January 2007, LJN: AZ7033
Vie d’Or, Amsterdam Court of Appeal, 29 April 2009, LJN: BI2717
Shell, Amsterdam Court of Appeal, 29 May 2009, LJN: BI5744
Vedior, Amsterdam Court of Appeal, 15 July 2009, LJN: BJ2691
103
Converium, Amsterdam Court of Appeal, 17 January 2012, LJN: BV1026
DES II, Amsterdam Court of Appeal, 24 June 2014, LJN: 2372
DSB Bank, Amsterdam Court of Appeal, 4 November 2014, LJN: 4560
Secondary sources
Books and articles
A. Alemanno & A.-L. Sibony (eds), Nudge and the Law: A European Perspective (Hart
Publishing, 2015).
T. Arons & W.H. van Boom, 'Beyond Tulips and Cheese: Exporting Mass Securities Claim
Settlements from The Netherlands' (2010) European Business Law Review Vol. 21(6), 857.
D. Ashton & D. Henry, Competition Damages Actions in the EU: Law and Practice (Edward
Elgar Publishing 2013).
L. Athanassiou, ‘Collective Redress and Competition Policy’ in A. Nuyts & N.E. Hatzimihail
(eds), Cross-Border Class Actions: The European Way (Sellier European Law Publishers
2014).
I. Ayres & J. Braithwaite, Responsive Regulation: Transcending the Deregulation Debate
(Oxford University Press 1992).
R. Baldwin, M. Cave & M. Lodge, Understanding Regulation: Theory, Strategy, and Practice
(Oxford University Press 2012).
R. Baldwin & J. Black, ‘Really Responsive Regulation’ (2008) Modern Law Review Vol.
71(1), 59.
I. Bantekas, An Introduction to International Arbitration (Cambridge University Press 2015).
H. Bergman & D.D. Sokol, ‘The Air Cargo Cartel: Lessons for Compliance’ in Beaton-Wells
& Tran (eds), Anti-Cartel Enforcement In A Contemporary Age: Leniency Religion (Hart
Publishing 2015).
S. Bhattacharya & R. Van den Bergh, ‘The Contribution of Management Studies to
Understanding Firm Behaviour and Competition Law’ (2014) World Competition Vol. 37(4),
517.
P. Billiet (ed), Class Arbitration in the European Union (Maklu 2013).
J.H.J. Bourgeois & S. Strievi, ‘EU Competition Remedies in Consumer Cases: Thinking Out
of the Shopping Bag’ (2010) World Competition Vol. 33(2), 241.
104
M. Bredenoord-Spoek, ‘The Dutch Competition Authority Considers Cooperation and Joint
Selling of Network Services for PIN-transactions to be an Infringement of the Cartel
Prohibition (Interpay)’ (2005) e-Competitions Bulletin Excessive Prices Art. No. 35979.
O.W. Brouwer, ‘Antitrust Settlements in the Netherlands: A Useful Source of Inspiration?’ in
C.D. Ehlermann & M. Marquis (eds), European Competition Law Annual 2008: Antitrust
Settlements under EC Competition Law (Hart Publishing 2010).
C.J. Buckner, ‘Due Process in Class Arbitration’ (2006) Florida Law Review Vol. 58, 185.
M. Cappelletti, ‘Alternative Dispute Resolution Processes Within the Framework of the
World Wide Access to Justice Movement’ (1993) Modern Law Review Vol. 56(3), 282.
H. Collins, ‘Regulating Contract Law’ in C. Parker and others (eds), Regulating Law (Oxford
University Press 2004).
H. Collins, Regulating Contracts (Oxford University Press 2002).
D.A. Crane, 'Optimizing Private Antitrust Enforcement' (2010) Vanderbilt Law Review Vol.
63(3), 673.
N. Creutzfeldt, ‘How Important is Procedural Justice for Consumer Dispute Resolution? A
Case Study of an Ombudsman Model for European Consumers’ (2014) Journal of Consumer
Policy Vol. 37, 527.
P. Cserne, ‘Making Sense of Nudge-Scepticism: The Challenges to EU Law’s Learning from
Behavioural Sciences’ in A. Alemanno & A.-L. Sibony (eds), Nudge and the Law: A
European Perspective (Hart Publishing 2015).
H. Eidenmüller (ed), Regulatory Competition in Contract Law and Dispute Resolution (Hart
Publishing 2013).
Y. Feldman & H.E. Smith, ‘Behavioral Equity’ (2014) Journal of Institutional and Theoretical
Economics Vol. 170, 137.
R. Fentiman, ‘Recognition, Enforcement and Collective Judgments’ in A. Nuyts & N.E.
Hatzimihail (eds), Cross-Border Class Actions: The European Way (Sellier European Law
Publishers 2014).
O.M. Fiss, ‘Against Settlement’ (1984) Yale Law Journal Vol. 93, 1073.
D. Geradin, 'Antitrust Compliance Programmes and Optimal Antitrust Enforcement: a Reply
to Wouter Wils' (2013) Journal of Antitrust Enforcement Vol. 1(2), 325.
L. Gorywoda, ‘The Emerging EU Legal Regime for Collective Redress: Institutional
Dimension and Its Main Features’ in A. Nuyts & N.E. Hatzimihail (eds), Cross-Border Class
Actions: The European Way (Sellier European Law Publishers 2014).
N. Gunningham, P. Grabosky & D. Sinclair, Smart Regulation: Designing Environmental
Policy (Clarendon 1998).
105
N. Gunningham, ‘Strategizing Compliance and Enforcement: Responsive Regulation and
Beyond’ in C. Parker & V.L. Nielsen (eds), Explaining Compliance: Business Responses to
Regulation (Edward Elgar Publishing 2011).
N. Gunningham, ‘Enforcement and Compliance Strategies’ in R. Baldwin, M. Cave & M.
Lodge (eds), The Oxford Handbook of Regulation (Oxford University Press 2010).
C. Hodges, ‘US Class Actions: Promise and Reality’ in H.-W. Micklitz & A. Wechsler (eds),
The Transformation of Enforcement: European Economic Law in Global Perspective (Hart
Publishing 2016).
C. Hodges & N. Creutzfeldt, 'Transformations in Public and Private Enforcement' in H.-W.
Micklitz & A. Wechsler (eds), The Transformation of Enforcement: European Economic Law
in Global Perspective (Hart Publishing 2016).
C. Hodges, Law and Corporate Behaviour: Integrating Theories of Regulation, Enforcement,
Compliance and Ethics (Hart Publishing 2015).
C. Hodges, ‘Mass Collective Redress: Consumer ADR and Regulatory Techniques’ (2015)
European Review of Private Law Vol. 23(5), 829.
C. Hodges, ‘European Competition Enforcement Policy: Integrating Restitution and
Behaviour Control’ (2011) World Competition Vol. 34(3), 383.
C. Hodges, I. Benöhr & N. Creutzfeldt-Banda (eds), Consumer ADR in Europe (Hart
Publishing 2012).
C. Hodges & A. Stadler (eds), Resolving Mass Disputes: ADR and Settlement of Mass Claims
(Edward Elgar Publishing 2013).
C. Hodges, The Reform of Class and Representative Actions in European Legal Systems: A
New Framework for Collective Redress in Europe (Hart Publishing 2008).
R. Hollander-Blumoff & T.R. Tyler, ‘Procedural Justice and the Rule of Law: Fostering
Legitimacy in Alternative Dispute Resolution’ (2011) Journal of Dispute Resolution Vol. 1, 1.
I. Hunter & L. Flannery, ‘Class Action and Arbitration Procedures - United Kingdom’ in P.
Billiet (ed), Class Arbitration in the European Union (Maklu 2013).
M. Ioannidou, Consumer Involvement in Private EU Competition Law Enforcement (2nd edn,
Oxford University Press 2015).
Z. Juska, ‘Obstacles in European Competition Law Enforcement: A Potential Solution from
Collective Redress’ (2014) European Journal of Legal Studies Vol.7(1), 125.
P. Kalbfleisch, ‘The Dutch Experience with Plea Bargaining/Direct Settlement’ in C.D.
Ehlermann & M. Marquis (eds), European Competition Law Annual 2008: Antitrust
Settlements under EC Competition Law (Hart Publishing 2010).
106
S.E. Keske, Group Litigation in European Competition Law: A Law and Economics
Perspective (Intersentia 2010).
A.S. Kim, ‘Rent-a-judges and the Cost of Selling Justice’ (1994) Duke Law Journal Vol. 44,
166.
J.S. Kortmann & M.C. Buiten, ‘Private Enforcement and Collective Redress in European
Competition Law’ (2015) FIDE 2016 Questionnaire.
B. Krans, ‘The Dutch Act on Collective Settlement of Mass Damages’ (2014) Global
Business & Development Law Journal Vol. 27(2), 281.
E.C. Kurucz, B.A. Colbert & D. Wheeler, ‘The Business Case for Corporate Social
Responsibility’ in A. Crane (ed), The Oxford Handbook of Corporate Social Responsibility
(Oxford University Press 2008).
E. Lachnit, ‘Compliance Programmes in Competition Law: Improving the Approach of
Competition Authorities’ (2014) Utrecht Law Review Vol. 10(5), 31.
E. Lachnit, ‘Alternative Enforcement of Competition Law – Balancing Legal Requirements in
Practice’ (2013) RENFORCE Working Paper, Utrecht Centre for Regulation and
Enforcement in Europe.
I. Lianos, P. Davis & P. Nebbia, Damages Claims For the Infringement of EU Competition
Law (Oxford University Press 2015).
M. Lodge & K. Wefrich, Managing Regulation: Regulatory Analysis, Politics and Policy
(Palgrave Macmillan 2012).
Lord Woolf, ‘Access to Justice: Final Report to the Lord Chancellor on the Civil Justice
System in England and Wales’ (1996) Her Majesty's Stationery Office.
J.R. Macey & G.P. Miller, ‘Judicial Review of Class Action Settlements’ (2009) Journal of
Legal Analysis Vol. 1(1), 167.
A.R. Mele & P. Rawling (ed), The Oxford Handbook of Rationality (Oxford University Press
2004).
C. Menkel-Meadow, ‘Whose Dispute Is it Anyway?: A Philosophical and Democratic
Defense of Settlement (In Some Cases)’ (1995) Georgia Law Review Vol. 83, 2663.
H.-W. Micklitz & A. Wechsler (eds), The Transformation of Enforcement: European
Economic Law in Global Perspective (Hart Publishing 2016).
R. Mulheron, 'A Spotlight on the Settlement Criteria Under the United Kingdom's New
Competition Class Action' (2016) Civil Justice Quarterly Vol. 35(1), 14.
R. Mulheron, ‘Third Party Funding and Class Actions Reform’ (2015) Law Quarterly Review
Vol. 131(Apr), 291.
107
R. Mulheron, ‘Cy-Près Damages Distributions in England: A New Era for Consumer
Redress’ (2009) European Business Law Review Vol. 20(2), 307.
R. Mulheron, ‘Reform of Collective Redress in England and Wales: A Perspective of Need. A
Research Paper for Submission to the Civil Justice Council of England and Wales’ (2008)
Civil Justice Council of England and Wales.
A. Ottow, Market and Competition Authorities: Good Agency Principles (Oxford University
Press 2015).
M. Palmer, ‘Formalisation of Alternative Dispute Resolution Processes: Some Socio-Legal
Thoughts’ in Zekoll, Bälz & Amelung (eds), Formalisation and Flexibilisation in Dispute
Resolution (Brill 2014).
C. Parker, The Open Corporation: Effective Self-Regulation and Democracy (Cambridge
University Press 2002).
C. Parker & V.L. Nielsen, Explaining Compliance: Business Responses to Regulation
(Edward Elgar Publishing 2011).
M. Pel, ‘Regulation of Dispute Resolution in the Netherlands: Does Regulation Support or
Hinder the Use of ADR?’ in F. Steffek & H. Unberath (eds), Regulating Dispute Resolution:
ADR and Access to Justice at the Crossroads (Hart Publishing 2013).
A. Renda, ‘Private Antitrust Damages Actions in the EU’ in H.-W. Micklitz & A. Wechsler
(eds), The Transformation of Enforcement: European Economic Law in Global Perspective
(Hart Publishing 2016).
S. Roberts & M. Palmer, Dispute Processes: ADR and the Primary Forms of DecisionMaking (Cambridge University Press 2005).
B. Rodger, ‘Private Enforcement of Competition Law, the Hidden Story: Part II: Competition
Litigation Settlements in the UK, 2008-2012’ (2015) Global Competition Litigation Review
Vol. 8(3), 89.
B. Rodger, ‘The Consumer Rights Act 2015 and Collective Redress for Competition Law
Infringements in the UK: a Class Act?’ (2015) Journal of Antitrust Enforcement Vol. 3(2),
258.
B. Rodger, ‘Competition Law Compliance: the CMA 2015 Study, Compliance Rationales and
the Need for Increased Compliance, Professionalism and Education’ (2015) European
Competition Law Review Vol. 36(10), 423.
B. Rodger, ‘Private Enforcement of Competition Law, the Hidden Story: Competition
Litigation Settlements in the United Kingdom, 2000-2005’ (2008) European Competition
Law Review Vol. 29(2), 96.
M. Shapiro, Courts: A Comparative and Political Analysis (The University of Chicago Press
1981).
108
M. Shapiro, ‘Class Actions: The Class as Party and Client’ (1998) Notre Dame Law Review
Vol. 73(4), 913.
S.S. Simpson & M. Rorie, ‘Motivating Compliance: Economic and Material Motives for
Compliance’ in C. Parker & V.L. Nielsen (eds), Explaining Compliance: Business Responses
to Regulation (Edward Elgar Publishing 2011).
L.B. Solum, ‘Procedural Justice’ (2004) Southern California Law Review Vol. 78, 181.
A. Stadler, ‘Mass Damages in Europe – Allocation of Jurisdiction – Cross-border
Multidistrict Litigation’ in W.H. van Boom & G. Wagner (eds), Mass Torts in Europe: Cases
and Reflections (Walter de Gruyter 2014).
J. Stapleton, ‘Regulating Torts’ in C. Parker and others (eds), Regulating Law (Oxford
University Press 2004).
J. Steele & W.H. van Boom (eds), Mass Justice: Challenges of Representation and
Distribution (Edward Elgar Publishing 2011).
S.I. Strong, Class, Mass, and Collective Arbitration in National and International Law
(Oxford University Press 2013).
S.I. Strong, ‘From Class to Collective: The De-Americanization of Class Arbitration’ (2010)
Arbitration International Vol. 26(4), 493.
C.R. Sunstein, ‘Empirically Informed Regulation’ (2011) The University of Chicago Law
Review Vol. 78(4), 1349.
Z.S. Tang, Electronic Consumer Contracts in The Conflict of Laws (2nd edn, Hart Publishing
2015).
I. Tzankova & D. Hensler, ‘Collective Settlements in the Netherlands: Some Empirical
Observations’ in C. Hodges & A. Stadler (eds), Resolving Mass Disputes: ADR and
Settlement of Mass Claims (Edward Elgar Publishing 2013).
T.R. Tyler, ‘The Psychology of Self-Regulation: Normative Motivations for Compliance’ in
C. Parker & V.L. Nielsen (eds), Explaining Compliance: Business Responses to Regulation
(Edward Elgar Publishing 2011).
W.H. van Boom, ‘Collective Settlements of Mass Claims in the Netherlands’ (2009)
Rotterdam Institute of Private Law Accepted Paper Series.
H. van Lith, The Dutch Collective Settlements Act and Private International Law (Maklu
2011).
G. Wagner, ‘Mass Tort Resolution: Competition Between Jurisdictions and Mechanisms’ in
W. van Boom & G. Wagner (eds), Mass Torts in Europe: Cases and Reflections (Walter de
Gruyter 2014).
109
F. Weber, The Law and Economics of Enforcing European Consumer Law: a Comparative
Analysis of Package Travel and Misleading Advertising (Ashgate Publishing 2014).
F. Weber & C. Hodges, ‘The Netherlands’ in C. Hodges, I. Benöhr & N. Creutzfeldt-Banda
(eds), Consumer ADR in Europe (Hart Publishing 2012).
A. Wechsler & B. Tripkovic, ‘Conclusions’ in H.-W. Micklitz & A. Wechsler (eds), The
Transformation of Enforcement: European Economic Law in Global Perspective (Hart
Publishing 2016).
E.J. Weinrib, The Idea of Private Law (2nd edn, Oxford University Press, 2012).
R. Whish & D. Bailey, Competition Law (8th edn, Oxford University Press, 2015).
W. Wils, ‘Antitrust Compliance Programmes and Optimal Antitrust Enforcement’ (2013)
Journal of Antitrust Enforcement, Vol. 1(1), 52.
W. Wils, ‘The Relationship between Public Antitrust Enforcement and Private Actions for
Damages’ (2009) World Competition Vol. 32(1), 3.
W. Wils, The Optimal Enforcement of EC Antitrust Law (Kluwer Law International 2002).
Other secondary sources
Authority of Consumers and Markets, ‘Chris Fonteijn speech on compliance at International
Chamber of Commerce’ [in Dutch] (Toespraak Chris Fonteijn bij International Chamber of
Commerce over compliance), The Hague, 24 January 2014.
Authority of Consumers and Markets, ‘Behavioural Economics and Competition Policy’, The
Hague, June 2013.
Authority for Consumers and Markets, ‘Strategy Document’, The Hague, September 2013.
Authority for Consumers and Markets, ‘Press Release: NMa Reviews Fines Imposed on
Banks and Interpay’, The Hague, 22 December 2005.
American Arbitration Association, ‘Supplementary Rules for Class Arbitration’, 8 October
2003.
Ashurst, ‘Study on the Conditions of Claims for Damages in Case of Infringement of EC
Competition Rules’, Brussels, 31 August 2004.
Competition and Markets Authority, ‘Guidance on the Approval of Voluntary Redress
Schemes for Infringements of Competition Law’, London, 14 August 2015.
Competition and Markets Authority, ‘Competition Act 1998: Guidance on the CMA's
Investigation Procedures in Competition Act 1998 Cases’ (CMA8), March 2014.
110
Competition and Markets Authority, ‘How Small Businesses Can Comply with Competition
Law’ (OFT1330), March 2014.
Dutch Government, ‘The Netherlands' Response to the White Paper on Damages Actions for
Breach of the EC Antitrust Rules’, 2008.
Dutch Government, ‘Dutch Response to the Green Paper on Consumer Collective Redress’,
2008.
Dutch Minister of Economic Affairs, ‘Policy Rule of the Minister of Economic Affairs, no.
WJZ/14112617, on the Imposition of Administrative Fines by the Netherlands Authority for
Consumers and Markets’, 4 July 2014.
Dutch Minister of Security and Justice, ‘Dutch Draft Bill on Redress of Mass Damages in a
Collective Action’, 7 July 2014.
European Commission, ‘Antitrust: Commission proposes legislation to facilitate damage
claims by victims of antitrust violations’ IP/13/525, Brussels, 11 June 2013.
European Commission, ‘Compliance matters: What companies can do better to respect EU
competition rules’, Luxembourg, 2012.
European Commission, ‘Staff Working Document, Public Consultation: Towards a Coherent
European Approach to Collective Redress’ SEC (2011) 173 final.
European Commission, ‘Antitrust: Commission fines 11 air cargo carriers €799 million in
price fixing cartel’ IP/10/1487, Brussels, 9 November 2010.
European Commission, ‘DG SANCO Consultation Paper’, Brussels, May 2009.
European Commission, ‘White Paper on Damages Actions for Breach of EC Antitrust Rules’
COM (2008) 165 final.
European Commission, ‘Green Paper on Consumer Collective Redress’ COM (2008) 794
final.
European Commission, ‘Press Release of 27 November 2008, Green Paper on Consumer
Collective Redress – Questions and Answers’ MEMO/08/741.
European Commission, ‘Green Paper on Damages Actions for Breach of the EC Antitrust
Rules’ COM (2005) 672 final.
European Law Institute, ‘Statement of the European Law Institute on Collective Redress and
Competition Damages Claims’, 2014.
IFF Research, ‘UK Businesses’ Understanding of Competition Law: Prepared for the CMA’,
26 March 2015.
Independent Post and Telecommunications Authority, ‘OPTA Annual Report and Market
Monitor 2007’, The Hague, 27 May 2008.
111
Judicial Arbitration and Mediation Services, ‘Class Action Procedures’, 1 May 2009.
Lufthansa, ‘Annual Report 2014’, Frankfurt, March 2015.
Lufthansa, ‘Annual Report 2006’, Cologne, March 2007.
Office of Fair Trading, ‘OFT's Guidance as to the Appropriate Amount of a Penalty’
(OFT423), September 2012.
Office of Fair Trading, ‘Guidance: How Your Business Can Achieve Compliance with
Competition Law’ (OFT1341), June 2011.
Office of Fair Trading, ‘What Does Behavioural Economics Mean for Competition Policy?’
(OFT1224), March 2010.
Office of Fair Trading, ‘OFT Issues Final Decision and Imposes Penalties in Independent
Schools Investigation’, 24 November 2006.
Office of Fair Trading, ‘Enforcement: Incorporating the Office of Fair Trading’s guidance as
to the circumstances in which it may be appropriate to accept commitments’ (OFT407),
December 2004.
Oxera Consulting, ‘Behavioural Economics and its Impact on Competition Policy: A Practical
Assessment with Illustrative Examples from Financial Services’, May 2013.
UK Department for Business Innovation & Skill, ‘Protection of Small Businesses When
Purchasing Goods and Services: Call for Evidence’, March 2015.
UK Department for Business Innovation & Skill, ‘Competition Appeal Tribunal Rules of
Procedure: Review by the Right Honourable Sir John Mummery: Consultation’, February
2015.
UK Department for Business Innovation & Skill, ‘Competition Appeal Tribunal Rules of
Procedure: Government Response’, September 2015.
UK Department for Business Innovation & Skill, ‘Private Actions in Competition Law: A
Consultation on Options for Reform - Government Response’, January 2013.
UK Department for Business Innovation & Skill, ‘Private Actions in Competition Law: A
Consultation on Options for Reform’, April 2012.
UK Ministry of Justice, ‘The Government’s Response to the Civil Justice Council’s Report:
Improving Access to Justice through Collective Actions’, July 2009.
Volkswagen Investor Settlement Foundation, ‘Press Release: Dutch Settlement Foundation
Seeks to Resolve Securities Claims both for the Benefit of Volkswagen and its Investors’, 15
February 2016.
112